Auto-claude-code-research-in-sleep (ARIS ⚔️🌙)

💡 Use ARIS in Claude Code / Cursor / Trae as a skill-based workflow, or get the full experience with the standalone CLI — enjoy any way you like!

🤖 AI agents: Read AGENT_GUIDE.md instead — structured for LLM consumption, not human browsing.

🔥 ARIS-Code CLI — 独立安装版 · English | ⬇️ Download

📰 ARIS-Code v0.4.5 → v0.4.9 (2026-05) — Five-release polish sequence. New providers: DeepSeek V4 Pro / Xiaomi MiMo / Qwen 3.6 / Doubao / Custom OpenAI-compatible / DashScope Coding Plan (via native-tls switch). Reasoning + tool-use first class: reasoning_effort='xhigh' on the wire for o-series / gpt-5.5 / DeepSeek-thinking, reasoning_content cache+replay for Kimi / Moonshot / Xiaomi MiMo / DeepSeek-R1, thinking content blocks end-to-end. Skill helper subsystem rewrite: bundled helpers materialise into ~/.config/aris/cache/<version>/ at startup (no more cwd pollution), SkillOutput.helperReport JSON + 4-layer fallback chain, new integration-contract.md with 6 failure policies, inventory test + smoke regression guard. Critical bug fixes: PermissionMode::Prompt was silently allowing every tool (derived-Ord bug, every version pre-v0.4.6); hardcoded current_date = "2026-03-31" made models reject real post-cutoff data; Custom reviewer reset to gpt-5.5 every restart (setup menu option 9 vs 8 typo); third-party Anthropic-compat proxies hit missing field signature (#228); gpt-5.5 / o3 / o4 + tools 400 on OpenAI. Credits: @GetIT-Sunday, @Anduin9527, @GO-player-hhy, @Jxy-yxJ, @screw-44, @StevenUST.

<details><summary>Per-release details (v0.4.5 → v0.4.9)</summary>

v0.4.9 (2026-05-17) — Closes Codex v0.4.7 audit residuals (L1 TLS double-stack, L3 reasoning_cache compaction misalign, L4 reasoning replay unbounded). 2 new skills bundled (/figure-spec + /paper-illustration-image2 with scripts/ subdirs, new Layer 0b = $ARIS_CACHE_DIR/skills/<name>/scripts/); research_wiki.py promoted to shared tools/ (9+ callers); 5 more SKILL.md migrated to fallback chain.

v0.4.8 (2026-05-17) — Skill helper subsystem rewrite. Bundled helpers extract to ~/.config/aris/cache/<version>/ at startup; every Skill invocation surfaces helperReport JSON + 4-layer resolver preamble; /skills export copies helpers; new integration-contract.md with 6 failure policies; 8 shared helpers (arxiv/deepxiv/exa/S2/openalex/save_trace/verify_papers/verify_paper_audits) bundled; /research-lit + /deepxiv migrated. Plus 4 bug fixes: gpt-5.5+tools 400 on OpenAI; Custom reviewer reset; missing signature field (#228); --version Build date hardcoded.

v0.4.7 (2026-05-16) — DashScope Coding Plan 405 fixed (#159) via native-tls switch (#225); reasoning_content replay for all reasoning models (OpenAI o1/o3/o4 / DeepSeek-R1 etc.), not just Kimi (#226); 600+ lines dead code cleanup + rustyline dep removed + "Claw Code" → "ARIS-Code" rebrand.

v0.4.6 (2026-05-14) — 🚨 Two long-standing silent bugs fixed: PermissionMode::Prompt silently allowed every tool (derived-Ord bug); system prompt hardcoded current_date = "2026-03-31" made models reject post-cutoff data as future/prompt-injection. Plus Custom OpenAI-compatible provider (/setup option 11) with dynamic /models discovery (@Anduin9527 #221 + #222).

v0.4.5 (2026-05-13) — First-class reasoning-model support: thinking content blocks end-to-end (fixes #161) + reasoning_effort='xhigh' for GPT-5.5 / o1 / o3 / o4 / DeepSeek-thinking. DeepSeek V4 Pro + Xiaomi MiMo + Qwen 3.6 + Doubao in /setup (options 7-10). Object-style hooks parser. Default model bumped to Claude Opus 4.7 + GPT-5.5. REPL input hardening (multi-line wrap / Cmd+V paste / CJK boundary). GitHub Actions CI. Credits: @GO-player-hhy (#186), @Jxy-yxJ (#171), @GetIT-Sunday (#216 partial).

</details> <details><summary>Older versions</summary>

v0.4.4 (2026-04-20) — Setup UX + reviewer routing fixes (resolves #158, #162) | /setup no longer forces Bearer for Anthropic + custom URL | Provider-aware proxy URL hints | Stale state no longer leaks across provider switches | LlmReview smart fallback

v0.4.3 (2026-04-17) — Third-party Anthropic-compat proxy support (Bedrock etc.) | Skip beta flags that proxies reject | Propagate custom base URL for anthropic provider | Credit @screw-44

v0.4.2 (2026-04-17) — Auto-compaction corruption fix | Compaction summary preserved on OpenAI-compat executors | Shell-provided API keys no longer erased on launch

v0.4.1 (2026-04-15) — Plan mode (/plan) | Cooperative Ctrl+C interrupt | Auto-retry (429/5xx/network) | Research Wiki 📚 (persistent knowledge base) | Self-Evolution 🧬 (/meta-optimize) | Local models (LM Studio/Ollama) | 62 skills synced

v0.3.11 (2026-04-13) — Reviewer Anthropic-compatible mode (Claude via proxy)

v0.3.9 (2026-04-11) — Proxy/custom base URL (CCSwitch) | Local models (LM Studio/Ollama) | Windows (experimental)

v0.3.5 (2026-04-08) — Research Wiki (persistent papers/ideas/experiments/claims + relationship graph) | Meta-Optimize self-evolution (analyze logs → propose SKILL.md patches)

v0.3.0 (2026-04-03) — Multi-file memory index | Rich task system (TodoWrite) | /plan | Security hardening

v0.2.2 (2026-04-03) — /plan step-by-step planning | /tasks persistent tracking

v0.2.1 (2026-04-03) — Persistent Memory | Kimi K2.5 multi-turn fix | CJK cursor fix

v0.2.0 (2026-04-02) — Open source | Kimi + MiniMax + GLM support | Smart LlmReview routing | CI/CD

v0.1.0 (2026-04-02) — Initial release | Multi-executor & reviewer | 42 bundled skills

</details>

中文版 README | English

🌙 Let Claude Code do research while you sleep. Wake up to find your paper scored, weaknesses identified, experiments run, and narrative rewritten — autonomously.

🪶 Radically lightweight — zero dependencies, zero lock-in. The entire system is plain Markdown files. No framework to learn, no database to maintain, no Docker to configure, no daemon to babysit. Every skill is a single SKILL.md readable by any LLM — swap Claude Code for Codex CLI, OpenClaw, Cursor, Trae, Antigravity, Windsurf, or your own agent and the workflows still work. Fork it, rewrite it, adapt it to your stack.

💡 ARIS is a methodology, not a platform. What matters is the research workflow — take it wherever you go. 🌱

· · · · · · 💬 Join Community ·

Custom Claude Code skills for autonomous ML research workflows. These skills orchestrate cross-model collaboration — Claude Code drives the research while an external LLM (via Codex MCP) acts as a critical reviewer. 🔀 Also supports alternative model combinations (Kimi, LongCat, DeepSeek, etc.) — no Claude or OpenAI API required. For example, MiniMax-M2.7 + GLM-5 or GLM-5 + MiniMax-M2.7. 🤖 Codex CLI native — full skill set also available for OpenAI Codex. 🖱️ Cursor — works in Cursor too. 🖥️ Trae — ByteDance AI IDE. 🚀 Antigravity — Google's agent-first IDE. 🆓 Free tier via ModelScope — zero cost, zero lock-in.

💭 Why not self-play with a single model? Using Claude Code subagents or agent teams for both execution and review is technically possible, but tends to fall into local minima — the same model reviewing its own patterns creates blind spots.

Think of it like adversarial vs. stochastic bandits: a single model self-reviewing is the stochastic case (predictable reward noise), while cross-model review is adversarial (the reviewer actively probes weaknesses the executor didn't anticipate) — and adversarial bandits are fundamentally harder to game.

💭 Why two models, not more? Two is the minimum needed to break self-play blind spots, and 2-player games converge to Nash equilibrium far more efficiently than n-player ones. Adding more reviewers increases API cost and coordination overhead with diminishing returns — the biggest gain is going from 1→2, not 2→4.

Claude Code's strength is fast, fluid execution; Codex (GPT-5.4 xhigh) is slower but more deliberate and rigorous in critique. These complementary styles — speed × rigor — produce better outcomes than either model talking to itself.

🧿 Want the strongest possible reviewer? Add — reviewer: oracle-pro to any skill to route reviews through GPT-5.4 Pro via Oracle MCP. Pro-level reasoning for proof verification, experiment auditing, and final stress tests. Works with API key or free browser mode. Setup →

🎯 More Than Just a Prompt

These are full pipelines — you can also use each workflow independently. Already have an idea? Skip to Workflow 1.5. Have results? Jump to Workflow 3. Got reviews? Jump to Workflow 4. Want persistent memory? Enable Research Wiki. See Quick Start for all commands and Workflows for the full breakdown.

Basic mode — give ARIS a research direction, it handles everything:

/research-pipeline "factorized gap in discrete diffusion LMs"

🔥 Targeted mode — got a paper you want to improve? Give ARIS the paper + the code:

/research-pipeline "improve method X" — ref paper: https://arxiv.org/abs/2406.04329, base repo: https://github.com/org/project

ARIS reads the paper → finds its weaknesses → clones the codebase → generates ideas that specifically fix those weaknesses with that code → runs experiments → writes your paper. Like telling a research assistant: "read this paper, use this repo, find what's missing, and fix it."

Mix and match: ref paper only = "what can be improved?", base repo only = "what can I build with this code?", both = "improve this paper using this code."

🔥 Rebuttal mode — reviews just dropped? Don't panic. ARIS reads every concern, builds a strategy, and drafts a rebuttal that's grounded, structured, and under the character limit:

/rebuttal "paper/ + reviews" — venue: ICML, character limit: 5000

Three safety gates — rebuttal will NOT finalize if any fails:

• 🔒 No fabrication — every claim maps to paper/review/user-confirmed result
• 🔒 No overpromise — every promise is user-approved
• 🔒 Full coverage — every reviewer concern is tracked

Two outputs: PASTE_READY.txt (exact char count, paste to venue) + REBUTTAL_DRAFT_rich.md (extended version for manual editing).

After acceptance — your paper is in, now prepare the presentation:

/paper-slides "paper/"     # → Beamer PDF + PPTX + speaker notes + Q&A prep
/paper-poster "paper/"     # → A0/A1 poster PDF + editable PPTX + SVG

💡 From idea to paper to podium — one toolchain. 🌱

📢 What's New

• 2026-05-17 — 🛠 Tools-stability roadmap (Phase 1+2+3) complete (closes #176 / #177 / #178). Community reported that helper scripts weren't propagating into user projects after install_aris.sh. Phase 1 — every SKILL.md caller of the 10 canonical helpers now resolves via the strict-safe 3-layer chain .aris/tools/ → tools/ → $ARIS_REPO/tools/ documented in integration-contract.md §2 (which also defines 5 failure policies A/B/C/D1/D2/E). Phase 2 — new advisory CI lint catches hardcoded python3 tools/foo.py patterns in PR-modified SKILL.md (advisory only, never fails CI). Phase 3 — three single-owner helpers (figure-spec, paper-illustration-image2, experiment-queue) moved into their SKILL's scripts/ subdirectory; owner SKILLs use Layer 0 ${CLAUDE_SKILL_DIR}/scripts/ ahead of the canonical chain; legacy tools/ paths retained as os.execv Python forwarding shims. ⚠️ Existing users: no action needed — legacy tools/ entries are now shims. If you haven't run install_aris.sh since 2026-04-30, one idempotent rerun catches everything up.
• 2026-05-14 — 🩹 /paper-plan + /paper-write learn GAP_REPORT.md + <!-- DATA_NEEDED --> discipline (#217). When — style-ref: is set and the user's project has structural assets (figures/, results/, NARRATIVE_REPORT.md, etc.), /paper-plan emits a Gap Report mapping the exemplar's section topology + density (from style_profile.md) against your actual assets — surfacing slots you have no evidence to fill (e.g., "exemplar has 3×4 ablation table, you have no ablation data"). Then /paper-write writes <!-- DATA_NEEDED: <Slot ID> — <description> --> HTML comments instead of fabricating content at missing slots — invisible in the compiled PDF, grep-friendly for human triage / /experiment-bridge follow-up. Narrow carve-out from the "no placeholders" rule, scoped to GAP_REPORT-listed slots only. Original idea by @zhangpelf.

• 2026-05-14 — ⚙️ Default reviewer model: gpt-5.4 → gpt-5.5 across ~30 SKILL.md REVIEWER_MODEL defaults. Codex MCP has routed gpt-5.5 as the default since 2026-04-24; this catches the docs up to runtime. ⚠️ Behavior changes: (a) .aris/traces/* JSONs from prior runs are not reproducible — re-runs on 5.5 may emit different WARN/FAIL verdicts on borderline cases (reviewer-quality lift, not regression). (b) ChatGPT Plus/Pro monthly quotas drain faster under heavy use. Fallback: pass — reviewer-model: gpt-5.4 per invocation, or pin REVIEWER_MODEL = gpt-5.4 per skill. Oracle Pro tier (routed via — reviewer: oracle-pro) is a separate path and unaffected.

• 2026-05-13 — 🔍 tools/verify_papers.py + Pre-Search Verification Protocol — anti-hallucination filter for literature-facing skills. New helper does 3-layer fallback verification (arXiv batch API up to 40 IDs/request → CrossRef DOI lookup → Semantic Scholar fuzzy title match, default 0.6 word-overlap) and emits 4-state per-paper status (verified / unverified / verify_pending / error) plus a top-level verdict aligning with assurance-contract.md (PASS / WARN / BLOCKED / ERROR). Transient failures (5xx, timeouts, 429) are tagged verify_pending and excluded from the hallucination rate so network blips don't get conflated with fabricated references. Per-project cache at <project>/.aris/cache/verify_papers.json with 30-day TTL; canonical key priority arxiv:{id_without_version} → doi:{lowercase} → title:{sha1[:16]}. New Pre-Search Verification Protocol subsection in shared-references/citation-discipline.md makes the split explicit: this protocol is the fast filter between SEARCH (Step 1) and full VERIFY (Step 2); /citation-audit and /paper-claim-audit remain the submission-time audit gates and are not replaced. /research-lit gets a mandatory Step 1.5: Verify Candidate Papers calling the helper; /idea-creator and /novelty-check add a Key Rule reference for cited Closest Prior Work / landscape entries. Unverified papers are retained in output tagged [UNVERIFIED] (retention-over-silent-removal) so search-quality issues stay visible. Set ARIS_VERIFY_EMAIL in your shell to lift CrossRef to the polite-pool rate. Original signal from @YiwenZhu77 in #120 — landed via clean reimplementation rather than direct merge (PR was 5 weeks old + scope creep into figure-style).

• 2026-05-06 — 🎤 /paper-talk workflow + /slides-polish skill — end-to-end conference talk pipeline. /paper-talk orchestrates paper → slide outline → Beamer + PPTX → per-page polish → assurance audits → final report (sister to /paper-writing, /paper-poster); composes /paper-slides, /slides-polish, plus /paper-claim-audit + /citation-audit when assurance: conference-ready. /slides-polish is the post-generation visual pass: per-page Codex review against a reference PDF + a fix-pattern catalog (PPTX font scaling 1.5-1.8× for projector-readable size, text-frame resize after font bump, banner-as-tcolorbox, italic style leak guard, em-dash spacing, Chinese EA font hint via PingFang SC, anonymity placeholder discipline). Assurance ladder draft / polished (default) / conference-ready is independent from the effort axis; effort: lite, assurance: conference-ready is legal and means "fast pipeline, every audit must emit verdict before final". Phase 4 staging adapter materializes slide text + speaker notes + talk script as a synthetic paper directory (.aris/paper-talk/audit-input/sections/*.tex + symlinked .bib / results/ / figures/) so the existing audits run with their paper-shaped contracts and emit 6-state JSON verdicts per shared-references/assurance-contract.md.

• 2026-05-05 — 🔁 /resubmit-pipeline — Workflow 5: text-only resubmit across venues (#208). Port a polished paper from one venue to another under hard constraints (no new experiments, no bib edits, no framework changes, never overwrite prior submissions). 5 phases: physical isolation → 5-layer anonymity check → audits (proof / claim / citation --soft-only) → microedits via /auto-paper-improvement-loop --edit-whitelist with per-round diff gate → adversarial gate via /kill-argument → final compile + Overleaf push via /overleaf-sync. Two prerequisite SKILL upgrades shipped in the same PR: /auto-paper-improvement-loop --edit-whitelist <path> (YAML schema with allowed/forbidden paths + forbidden_operations like new_cite / new_theorem_env / numerical_claim, forbidden_deletions, requires_user_approval_for, max_edits_per_round) and /citation-audit --soft-only (translates KEEP/FIX/REPLACE/REMOVE verdicts to text-rewrite proposals when bib is frozen; hallucinated citations get drop_cite_in_body_only action). Master RESUBMIT_REPORT.json ledger per shared-references/assurance-contract.md; 7-verdict failure mode table including USER_DECISION runtime state.

• 2026-05-05 — 🗡 /kill-argument — adversarial Attack-Adjudication review for theory papers (#206). Two fresh codex 5.5 + xhigh threads: Thread 1 writes the strongest 200-word rejection memo a senior area chair would produce; Thread 2 (independent adjudicator, NOT defender) reads the current paper and classifies each rejection point as answered_by_current_text / partially_answered / still_unresolved with file:line evidence. Output: KILL_ARGUMENT.{md,json}, detect-only. Integrated as Phase 5.6 of /paper-writing (between claim-audit and citation-audit) and as the canonical implementation called from /auto-paper-improvement-loop Step 5.5 — replaces inline prompt in both places. Mandatory at assurance: submission for theory-heavy / scope-heavy papers; emits NOT_APPLICABLE for empirical papers without scope claims. Audit JSON is verify_paper_audits.sh-compatible (full schema per shared-references/assurance-contract.md, 6-state verdict). Catches the failure mode score-based reviews miss: when every local component is correct (numbers match, cites resolve, theorems prove) but the paper still oversells what it actually establishes.

• 2026-05-04 — 🪲 /research-wiki and 8 caller skills now resolve helper via fallback chain (#204). Bug: after bash tools/install_aris.sh the helper lives at .aris/tools/research_wiki.py (symlink), but skills hard-coded tools/research_wiki.py and silently failed when invoked — research-wiki/ stayed empty across full W1 runs. Fix: 3-layer chain (.aris/tools/ → tools/ → $ARIS_REPO/tools/) codified in shared-references/wiki-helper-resolution.md. The manual-copy workaround at <project>/tools/research_wiki.py is layer 2, so users who cp-installed the helper as a temporary fix continue to work. ⚠️ Existing users: rerun bash tools/install_aris.sh once — also picks up a separate Python 3.9 ImportError fix in the helper.

• 2026-05-03 — 🎨 Opt-in — style-ref: <source> for writer-side skills (#202). /paper-{plan,write,writing,illustration,poster,slides}, /grant-proposal, and /auto-paper-improvement-loop accept an optional — style-ref: <source> argument that mimics a reference paper's structural style (section ordering, theorem/figure density, sentence cadence, citation style) without copying its prose, claims, or terminology. Sources: local .tex dir/file, local PDF, arXiv id (2501.12345 or arxiv:2501.12345), HTTP/HTTPS URL. Overleaf URLs/IDs are rejected — clone via /overleaf-sync setup <id> first. Default OFF; existing behavior unchanged when the flag is absent. Reviewer / auditor sub-skills (/proof-checker, /paper-claim-audit, /citation-audit, the improvement-loop reviewer) never see the style ref — cross-model review independence preserved. ⚠️ Existing ARIS users: the helper ships at tools/extract_paper_style.py, distributed via the .aris/tools symlink (install_aris.sh Phase 0, added in #192). Re-run bash tools/install_aris.sh once to refresh the symlink and pick up the helper. Manual fallback: cp <ARIS-repo>/tools/extract_paper_style.py <your-project>/tools/. Without either, the writer skill aborts with a clear error pointing here.

• 2026-05-02 — 🪨 Community spotlight: rosetta by @SyntaxSmith. Programmatic access to ChatGPT Pro / gpt-5.5-pro / DeepResearch from Node, via Chrome CDP Fetch interception + WebSocket second-leg streaming; ships an MCP server for Claude Code / Codex / Cline. Alternative implementation path to Oracle MCP for ARIS users invoking — reviewer: oracle-pro — same target capability (Pro-tier reviewer), different mechanics. Indexed under Awesome Community Skills & Extensions. 🌟 if you're using it!

• 2026-05-02 — 💎🧿 Model & MCP routing updates. (a) /gemini-search default bumped to gemini-3-pro-preview (strongest Gemini, out-of-box). ⚠️ Action required: requires gemini-cli v0.40+ (run gemini --version; upgrade with npm i -g @google/gemini-cli if older). Legacy override: /gemini-search "topic" — model: gemini-2.5-pro. Other overrides: gemini-3-flash-preview (faster), auto-gemini-3 (load-routed). (b) /idea-discovery Phase 1 now includes Gemini in its literature survey by default (#199) — auto-injects — sources: all, gemini into /research-lit unless the user passed an explicit — sources:; graceful skip if gemini-cli not installed. (c) Oracle MCP upstream PR queue (steipete/oracle/pulls) is the first triage stop when invoking — reviewer: oracle-pro (especially o3-deep-research / gpt-5.5-pro) — ARIS does not vendor Oracle MCP; check upstream first if behavior surprises you (reviewer-routing.md)

• 2026-05-02 — 🛠️🔗 Tools-infrastructure migration started. (a) install_aris.sh creates optional .aris/tools symlink (#192, closes #174) — Phase 0 of the 4-step tools-stability plan (#174 → #176 → #177 → #178); idempotent, zero impact until rerun. (b) /experiment-queue orchestration paths repaired (#193) — first real user of the symlink; 7 cascading bugs fixed via 3 rounds of Codex MCP gpt-5.5 xhigh audit. Pure prose + docstring; queue_manager.py logic untouched. Windows install_aris.ps1 parallel update tracked as follow-up

• 2026-05-02 — 🔬 Three new opt-in audit flags via fast-path delegated-agent workflow (#187, #188, #189). /citation-audit --uncited surfaces bib entries with no \cite{} reference (detect-only). /proof-checker --deep-fix adds a repair-grade plan to the Phase 1 reviewer prompt (corrected statement / patch plan / closure tests + Schur/quadratic-form algebra sanity). /proof-checker --restatement-check adds Phase 3.6 cross-location theorem drift detection (6 drift signatures). Zero behavior change when flags unset. Plus doc PRs #190 (thread-policy) + #191 (auto-loop xref). Delegated-agent + maintainer-fixup pattern; Codex MCP gpt-5.5 xhigh review caught 6+ blockers

• 2026-05-01 — 🔍 Gemini + OpenAlex literature sources for /research-lit (#175, community contribution by @stdAri). Two opt-in sources: /gemini-search (AI-driven discovery via jamubc/gemini-mcp-tool MCP) and /openalex (250M+ work open citation graph, no API key). Triggered via — sources: gemini or — sources: openalex; zero behavior change when default all (both excluded). Maintainer fixups: corrected @google/gemini-cli npm name; added try/except ImportError + bash preflight for graceful OpenAlex skip when requests missing

• 2026-04-30 — 📝 /rebuttal per-reviewer thread mode + transferable patterns (SKILL.md). Adds VENUE_MODE (single_document | per_reviewer_thread) for OpenReview-style venues, reviewer_priority: pivotal routing, structural_distinction response mode, 5 reviewer-defensive heuristics, 2 Phase 5 lints, and severity-scaled stress rounds. Default VENUE_MODE = single_document keeps ICML-style behavior — zero change for existing users. Three rounds of cross-model review before/after merge

• 2026-04-30 — 🪞 Codex skill mirror rebuilt + dedicated install/update chain (#179, community contribution by @No-518). skills/skills-codex/ now mirrors all 67 mainline skills; replaces mcp__codex__codex reviewer path with Codex-native spawn_agent + send_input. New tools/install_aris_codex.sh + tools/smart_update_codex.sh handle project-local symlinks with manifest tracking. Anti-drift: tests/test_codex_skill_mirror.py + tests/test_codex_install_update.py (26 failure paths). Open discussion in #184

• 2026-04-24 — 🎨 /paper-illustration-image2 — Codex-native image generation as Phase 2b illustration backend (#166, community contribution by @kbr19-thu 清华). Uses ChatGPT Plus/Pro quota via local Codex app-server MCP bridge — no GEMINI_API_KEY required. Triggered by /paper-writing — illustration: codex-image2; default stays figurespec (zero behavior change). Async-only API, sandboxed writes to figures/ai_generated/, integration-contract-compliant helper. Marked experimental (Codex debug app-server is unstable upstream)

• 2026-04-21 — 📚 Research Wiki ingest actually works now (research_wiki.py, /research-wiki). Fixes user-reported bug where /research-wiki init left papers/ empty forever (ingest subcommand had no implementation; paper-reading skills had no wiki hook). New canonical python3 tools/research_wiki.py ingest_paper helper owns slugging / metadata fetch / dedup / page render; all 6 paper-reading skills wired to it. Manual backfill via sync --arxiv-ids or sync --from-file. Ships with integration-contract.md formalizing the six-component pattern every cross-skill integration must follow

• 2026-04-21 — 🛡️ Assurance Gate: — effort: beast | max now really runs mandatory audits (assurance-contract.md, tools/verify_paper_audits.sh). Fixes silent-skip of /proof-checker / /paper-claim-audit / /citation-audit at high effort. New assurance axis (draft | submission) independent from effort: lite / balanced → draft (zero behavior change), max / beast → submission. At submission the 3 audits emit a JSON artifact with 6-state verdict; paper-writing Phase 6 runs the external verifier as source of truth (non-zero exit blocks Final Report). SHA256 input hashing catches stale audits. Escape hatch: — effort: beast, assurance: draft

• 2026-04-20 — 🩹 Project install: flat layout + manifest tracking — fixes a real bug where the previous nested install (.claude/skills/aris/) hid skills from Claude Code's slash-command discovery (CC only scans one directory level). Anyone who ran install_aris.sh before this date was silently affected. New install_aris.sh creates one symlink per skill at .claude/skills/<name>, writes a versioned manifest to .aris/installed-skills.txt, and is re-runnable to reconcile new/removed upstream skills. Defense-in-depth: 13 safety rules (no-symlinked-parents, exact-target revalidation, slug regex, atomic same-dir manifest rename, no-overwrite-real-files, mkdir-based portable lock, ADOPT for crash recovery, …). Granular --adopt-existing / --replace-link flags replace the all-or-nothing --force. Migration paths: --from-old for legacy nested symlink, --migrate-copy keep-user|prefer-upstream for legacy nested copy. smart_update.sh --target-subdir .claude/skills/aris is now deprecated with a redirect to install_aris.sh. Stale-file bug in cp -r overlay also fixed (now rm -rf && cp -r for safe-update path)

• 2026-04-19 — 🔗 /overleaf-sync — two-way bridge between local ARIS paper directory and an Overleaf project via the official Overleaf Git bridge (Premium). Lets collaborators keep editing in the Overleaf web UI while ARIS audit/edit pipelines (/paper-claim-audit, /citation-audit, /auto-paper-improvement-loop) keep running locally. Sub-commands: setup (one-time, user-driven so the agent never sees the token) / pull (with diff-protocol — flags half-sentences, typos, claim/cite changes that should re-trigger audits) / push (with confirmation gate before writing to shared Overleaf state) / status (3-way divergence check). Token never touches the agent or any file — primed once into macOS Keychain via the user's terminal, then auth-free for all subsequent agent operations

• 2026-04-19 — 📚 /citation-audit — fourth and final layer of the evidence-and-claim assurance stack (experiment-audit → result-to-claim → paper-claim-audit → citation-audit). Fresh cross-family reviewer (gpt-5.4 via Codex MCP) with web/DBLP/arXiv lookup verifies every \cite{...} along three independent axes: existence (paper resolves at claimed arXiv ID/DOI/venue), metadata correctness (authors/year/venue/title match canonical sources), and context appropriateness (the cited paper actually establishes the claim it supports — the most diagnostic check). Per-entry verdicts: KEEP / FIX / REPLACE / REMOVE. Auto-integrated into Workflow 3 Phase 5.8 as the pre-submission bibliography gate. Empirical motivation: in a real submission run, several real papers were cited in contexts they did not actually support, and at least one entry shipped with author = "Anonymous" — none caught by metadata-only checks

• 2026-04-17 — 🔀 /experiment-queue integrated into Workflow 1.5 + research-pipeline — experiment-bridge Phase 4 Deploy now auto-routes by milestone job count: ≤5 jobs → /run-experiment, ≥10 jobs or phase dependencies → /experiment-queue (with OOM retry, stale-screen cleanup, wave-transition gating, crash-safe state). New --- batch: queue override for global force-queue mode. Large multi-seed sweeps from EXPERIMENT_PLAN.md (e.g., 36-cell N × seed × n_train grids) now get proper orchestration without manual queue invocation

• 2026-04-17 — 🔗 Project-local symlink install (resolves #118) — new recommended default install. bash tools/install_aris.sh auto-detects platform (Claude Code / Codex CLI), creates .claude/skills/aris or .agents/skills/aris symlink to the ARIS repo, adds a managed <!-- ARIS:BEGIN --> block to CLAUDE.md / AGENTS.md telling the agent to use only project-local skills, and records install metadata in .aris/skill-source.txt. Solves the skill collision problem when ARIS is mixed with Superpowers / OpenHands / other community packs in the same global skill directory. PowerShell version (install_aris.ps1) ships with junction support for Windows. smart_update.sh --target-subdir flag added for .agents/skills/aris (Codex) project-copy installs; symlinked installs now correctly refuse smart_update and direct users to git pull. Global install remains supported for power users

• 2026-04-16 — 🎨 /figure-spec — deterministic JSON→SVG renderer packaged as a first-class skill. Preferred default for architecture/workflow/pipeline/audit-cascade figures in papers. Shape-aware edge clipping (rect/circle/ellipse/diamond), self-loops, curved edges, multi-line labels with CJK width estimation. Editable vector output, reproducible (same spec → same SVG), no external API. Phase 2b in Workflow 3 restored: illustration: figurespec (new default) / gemini / mermaid / false — 4-way illustration selector with complementary strengths

• 2026-04-16 — ⚙️ /experiment-queue — SSH job queue for multi-seed/multi-config ML experiments. Designed from real 36-cell NeurIPS sweep pain points: OOM-aware retry with backoff, stale-screen cleanup, wave-transition race prevention, teacher→student phase dependencies, crash-safe scheduler that resumes from JSON state. Declarative grid specs expand automatically (e.g., N × seed × n_train → 36 jobs). Configurable conda_hook + gpu_free_threshold_mib for non-standard environments. Use for ≥10 jobs; /run-experiment stays for ad-hoc

• 2026-04-15 — 🛡️ Paper Writing Pipeline Hardening — 10 empirically-motivated patches from a real NeurIPS run. REVIEWER_BIAS_GUARD=true: every review round uses a fresh thread (codex-reply inflated 3→8/10). Reviewer Independence Protocol: no fix summaries to reviewer. Step 4.5 Restatement Regression Test: catches theorem drift across fix rounds. Step 5.5 Kill Argument Exercise: final-round adversarial attack/defense for theory papers. Location-aware overfull blocking. Theory Paper Consistency Pass in /paper-write. Enforced Bib Hygiene with DBLP/CrossRef validation. Phase 5.5 Mandatory Final Claim Audit as submission gate. Review Tracing Protocol: full prompt/response pairs saved to .aris/traces/ for reviewer-independence audit (review-tracing.md, save_trace.sh). Inspired by community contribution from @李傲龍

• 2026-04-15 — 🎨 FigureSpec Renderer v2 — deterministic JSON→SVG figure generation for academic papers. Shape-aware edge clipping (rect/circle/ellipse/diamond), self-loops, curved edges, multi-line labels with CJK width estimation, comprehensive validation (type checks, structure, palette). Went through 5 rounds of Codex review (3/10→7/10). All architecture and workflow diagrams in the ARIS tech report were generated with this pipeline. New --- mode: vector for /paper-illustration skill

• 2026-04-14 — 📋 /paper-claim-audit — zero-context paper-to-evidence verification. Fresh reviewer with NO prior context compares every number in the paper against raw result files. Catches rounding inflation, best-seed cherry-pick, config mismatch, delta errors, scope overclaim. Auto-integrated into Workflow 3 (Phase 4.7). Completes the 3-layer audit chain: /experiment-audit (code) → /result-to-claim (science) → /paper-claim-audit (reporting). 👁️ Visual PDF review also added to improvement loop — reviewer now sees compiled PDF, not just LaTeX source. Inspired by Hermes Agent

• 2026-04-13 — 🧿 GPT-5.4 Pro via Oracle — — reviewer: oracle-pro on any skill for the strongest available reviewer. API mode (fast) or browser mode (free). Supported on: /research-review, /auto-review-loop, /experiment-audit, /proof-checker, /rebuttal, /idea-creator, /research-lit. Default stays Codex xhigh. Not installed = zero impact. Setup →

• 2026-04-13 — 🔬 /proof-checker — rigorous mathematical proof verification via cross-model review. 20-category issue taxonomy, two-axis severity, side-condition checklists (DCT/MCT/Fubini/IFT/...), counterexample red team, proof-obligation ledger. Auto-integrated into Workflow 3: detects \begin{theorem} and runs before improvement loop. Complements /proof-writer

• 2026-04-10 — ⚡ Effort Levels — — effort: lite | balanced | max | beast. Controls work intensity across all skills: papers found, ideas generated, review rounds, writing depth. Codex reasoning stays xhigh always. beast = every knob to maximum for top-venue sprints. Default balanced = zero change for existing users. Details →

• 2026-04-10 — 🔎 DeepXiv integration — progressive paper retrieval via DeepXiv CLI. Opt-in: — sources: deepxiv or — sources: all, deepxiv. Staged reading: search → brief → head → section. pip install deepxiv-sdk to enable. Community contribution by @DreamEnding

• 2026-04-10 — 🛡️ /experiment-audit — cross-model experiment integrity verification. GPT-5.4 reads your eval scripts and results directly, checks for fake ground truth, self-normalized scores, phantom results, and scope inflation (#131, #57). Advisory — warns loudly, never blocks. /result-to-claim auto-reads audit if present. New experiment-integrity.md shared reference. The executor must never judge its own integrity.

• 2026-04-10 — 🧠 tools/smart_update.sh — intelligent skill updater. Compares local vs upstream, detects personal customizations (server paths, API keys), only updates safe skills. bash tools/smart_update.sh --apply

• 2026-04-10 — 🏆 Community paper: UAV-CC — first community paper with full PDF archived. UAV change captioning benchmark for IEEE TGRS by @wxx827. Stack: Claude Opus 4.6 + Codex 5.4 xhigh + Cursor. Papers now archived in community_papers/

• 2026-04-08 — 📚 /research-wiki — persistent research knowledge base inspired by Karpathy's LLM Wiki. Accumulates papers, ideas, experiments, and claims across the entire research lifecycle with typed relationships. Wiki-aware hooks in /research-lit (ingest papers), /idea-creator (read wiki + write ideas back), and /result-to-claim (update claim status + trigger re-ideation). Failed ideas become anti-repetition memory. ARIS now learns from its mistakes.

• 2026-04-05 — 🧬 /meta-optimize — outer-loop harness optimization for ARIS. Passively logs skill invocations, tool calls, failures, and parameter overrides via Claude Code hooks. Run /meta-optimize to analyze accumulated usage data and propose SKILL.md improvements — reviewer-gated, user-approved. Inspired by Meta-Harness (Lee et al., 2026). ARIS now optimizes itself.

• 2026-04-04 — 🔧 Codex Plugin deep integration — /codex:rescue now auto-invoked when experiments fail (Workflow 1.5) or LaTeX won't compile (Workflow 3). GPT independently diagnoses the bug before Claude retries — two AI debuggers are better than one. Optional: codex exec powers nightmare review, /codex:rescue powers auto-debug. Setup →

• 2026-04-03 — ☁️ Modal serverless GPU — no GPU? gpu: modal in CLAUDE.md, one command (modal run launcher.py), no SSH, no Docker, auto scale-to-zero. $30/month free tier — enough to try ARIS experiments without any hardware. pip install modal && modal setup and go. Community contribution by @zeyuzhangzyz

• 2026-04-03 — 🎮 Reviewer Difficulty Levels — medium (default, unchanged), hard (reviewer memory + debate protocol), nightmare (GPT reads repo directly via codex exec — Claude can't hide anything). — difficulty: nightmare for maximum stress test before submission

• 2026-03-30 — 🔥 Auto-debug & exhaust-before-surrender — experiment-bridge auto-diagnoses failures (OOM, import, CUDA, NaN) and retries up to 3×. Inspired by PUA

• 2026-03-30 — ☁️ Vast.ai GPU rental — gpu: vast auto-rents cheapest GPU. By @YIHONG-JIN. 🔧 MiniMax M2.7 upgrade by @octo-patch

• 2026-03-27 — 📄 IEEE venue support (9 families). 🔎 Semantic Scholar. By @ypd666

• 2026-03-26 — 📄 Document-based input — RESEARCH_BRIEF.md auto-detect

• 2026-03-24 — 📝 Workflow 4: /rebuttal — 7-phase pipeline, 3 safety gates

• 2026-03-23 — 🔧 /training-check, /result-to-claim, /ablation-planner integrated. 📦 compact mode. By @JingxuanKang & @couragec

• 2026-03-22 — 📋 Templates — input templates for every workflow. 📄 7 venue templates — CVPR, ACL, AAAI, ACM MM added. 🛡️ Anti-hallucination fix — Workflow 2 enforces DBLP → CrossRef → [VERIFY]. 🔗 base repo — clone a GitHub repo as base codebase (— base repo: https://github.com/org/project)

• 2026-03-22 — 🔍 Codex + Gemini review guide — Codex executes, Gemini reviews via local gemini-review MCP bridge. CN

• 2026-03-20 — 🚀 Antigravity adaptation guide — use ARIS skills in Google Antigravity (agent-first IDE). Community contribution by @PeppaPigw

• 2026-03-20 — 🖥️ Trae adaptation guide — use ARIS skills in Trae (ByteDance AI IDE). Community contribution by @Prometheus-cotigo. 🔢 formula-derivation — Community contribution by @Falling-Flower

• 2026-03-19 — 🖼️ paper-poster — Conference poster. Community contribution by @dengzhe-hou

• 2026-03-19 — 🔗 Workflow 1.5 upgraded — /experiment-bridge GPT-5.4 code review. 📊 W&B fix

• 2026-03-18 — 🎤 paper-slides + 🔁 Codex+Claude bridge + 🖱️ Cursor guide + 🤖 Codex CLI skills + 📝 grant-proposal + 🎨 paper-illustration (Gemini) + 📊 CitationClaw

• 2026-03-17 — 🔧 Git code sync + 🆓 ModelScope guide + parameter pass-through

• 2026-03-16 — 🔬 research-refine + experiment-plan — turn vague ideas into problem-anchored proposals with claim-driven experiment roadmaps. Now integrated into Workflow 1 (/idea-discovery). Community contribution by @zjYao36

• 2026-03-16 — 🇨🇳 Alibaba Coding Plan guide — one API key, 4 models (Kimi-K2.5 + Qwen3.5+ + GLM-5 + MiniMax-M2.7), dual-endpoint setup. Community contribution by @tianhao909

• 2026-03-15 — 🔀 Bring your own model! Any OpenAI-compatible API now works as reviewer via llm-chat MCP server. GLM, MiniMax, Kimi, LongCat, DeepSeek all tested — zero Claude or OpenAI API needed

• 2026-03-15 — 🐾 OpenClaw adaptation guide — use ARIS research workflows in OpenClaw without Claude Code slash skills

• 2026-03-15 — 📐 proof-writer — community skill for rigorous theorem proof drafting. 📚 Anti-hallucination citations — /paper-write now fetches real BibTeX from DBLP/CrossRef instead of LLM-generated entries — on by default, zero install

• 2026-03-14 — 📱 Feishu/Lark integration: three modes (off/push/interactive), mobile notifications for experiments, reviews, and checkpoints

• 2026-03-13 — 🛑 Human-in-the-loop: configurable AUTO_PROCEED checkpoints across all workflows. Full autopilot or step-by-step approval

• 2026-03-12 — 🔗 Zotero + Obsidian + local PDFs + arXiv/Scholar: multi-source literature search with cross-model novelty verification

• 2026-03-12 — 🚀 Three end-to-end workflows complete: one prompt → top-venue-style paper. /research-pipeline chains idea discovery → auto review → paper writing autonomously

• 2026-03-12 — 📝 /paper-writing workflow: narrative report → structured outline → figures → LaTeX → compiled PDF → 2-round auto-improvement (4/10 → 8.5/10)

🚀 Quick Start

# 1. Install skills
git clone https://github.com/wanshuiyin/Auto-claude-code-research-in-sleep.git
mkdir -p ~/.claude/skills/    # create if it doesn't exist (new Claude Code versions)
cp -r Auto-claude-code-research-in-sleep/skills/* ~/.claude/skills/

# 1b. Update skills (when upstream has new versions)
cd Auto-claude-code-research-in-sleep && git pull
bash tools/smart_update.sh          # dry-run: shows what's new/changed/safe
bash tools/smart_update.sh --apply  # apply: adds new + updates safe ones

# Optional Codex mirror managed project install
bash tools/install_aris_codex.sh ~/your-codex-project

# Managed Codex project update
cd Auto-claude-code-research-in-sleep && git pull
bash tools/install_aris_codex.sh ~/your-codex-project --reconcile

# Copied Codex installs only (not for projects installed by install_aris_codex.sh)
bash tools/smart_update_codex.sh --local ~/.codex/skills
bash tools/smart_update_codex.sh --local ~/.codex/skills --apply

# 2. Set up Codex MCP (for review skills)
npm install -g @openai/codex
codex setup                    # set model to gpt-5.5 when prompted
claude mcp add codex -s user -- codex mcp-server

# 3. Use in Claude Code
claude
> /idea-discovery "your research direction"  # Workflow 1 — be specific! not "NLP" but "factorized gap in discrete diffusion LMs"
> /experiment-bridge                         # Workflow 1.5 — have a plan? implement + deploy + collect results
> /auto-review-loop "your paper topic or scope"  # Workflow 2: review → fix → re-review overnight
> /paper-writing "NARRATIVE_REPORT.md"       # Workflow 3: narrative → polished PDF
> /rebuttal "paper/ + reviews" — venue: ICML    # Workflow 4: parse reviews → draft rebuttal → follow-up
> /resubmit-pipeline "paper/" — venue: NeurIPS  # Workflow 5: port a polished paper to a new venue (text-only, no new experiments)
> /paper-talk "paper/" — venue: ICLR            # Workflow 6: paper → Beamer + PPTX talk + speaker notes + assurance audits
> /research-pipeline "your research direction"  # Full pipeline: Workflow 1 → 1.5 → 2 → 3 end-to-end
> /research-wiki init                           # 📚 Enable persistent research memory (one-time)
> /meta-optimize                                # Meta: analyze usage logs → propose skill improvements

Give ARIS persistent memory across sessions. Papers, ideas, failed experiments — nothing is forgotten:

# In Claude Code:
> /research-wiki init                         # creates research-wiki/ in your project
# That's it. From now on, /research-lit auto-ingests papers, /idea-creator reads
# the wiki before brainstorming (and writes ideas back), /result-to-claim updates
# claim status. Failed ideas become anti-repetition memory for future ideation.

Run these in your normal terminal (not inside Claude Code) to enable passive usage logging:

# One-time setup in your project directory
mkdir -p .claude .aris/meta tools/meta_opt
cp Auto-claude-code-research-in-sleep/templates/claude-hooks/meta_logging.json .claude/settings.json
cp Auto-claude-code-research-in-sleep/tools/meta_opt/*.sh tools/meta_opt/
chmod +x tools/meta_opt/*.sh
# Then start Claude Code — hooks are active immediately
claude

Events are logged to both project-level (.aris/meta/events.jsonl) and global (~/.aris/meta/events.jsonl) logs. After 5+ workflow runs, run /meta-optimize to see data-driven improvement proposals. Use /meta-optimize --global to analyze trends across all your projects.

📝 Templates available! See templates/ for ready-to-use input templates for every workflow — research brief (Workflow 1), experiment plan (Workflow 1.5), narrative report (Workflow 3), paper plan (Workflow 3).

🔎 Optional: DeepXiv progressive retrieval

pip install deepxiv-sdk

Then use /deepxiv directly or opt into it from /research-lit with — sources: deepxiv or — sources: all, deepxiv.

🔎 Optional: Exa AI-powered web search

pip install exa-py
export EXA_API_KEY=your-key-here

Then use /exa-search directly or opt into it from /research-lit with — sources: exa or — sources: all, exa. Covers blogs, docs, news, and research papers with built-in content extraction.

🗑️ Uninstall: To remove ARIS skills without affecting your own personal skills:

cd Auto-claude-code-research-in-sleep && ls skills/ | xargs -I{} rm -rf ~/.claude/skills/{}

All pipeline behaviors are configurable via inline overrides — append — key: value to any command:

/research-pipeline "your topic" — AUTO_PROCEED: false                          # pause at idea selection gate
/research-pipeline "your topic" — human checkpoint: true                       # pause after each review round to give feedback
/research-pipeline "your topic" — sources: zotero, web                         # only search Zotero + web (skip local PDFs)
/research-pipeline "your topic" — sources: all, deepxiv                        # default sources plus DeepXiv progressive retrieval
/research-pipeline "your topic" — sources: all, exa                            # default sources plus Exa AI-powered web search
/research-pipeline "your topic" — arxiv download: true                         # download top arXiv PDFs during literature survey
/research-pipeline "your topic" — difficulty: nightmare                        # maximum adversarial review before submission
/research-pipeline "your topic" — effort: beast                               # all knobs to maximum — top-venue sprint
/research-pipeline "your topic" — effort: beast, reviewer: oracle-pro         # beast + GPT-5.4 Pro reviewer — ultimate mode
/research-pipeline "your topic" — effort: lite                                # quick exploration, save tokens
/research-pipeline "your topic" — effort: max, review_rounds: 3               # max effort but cap review at 3 rounds
/research-pipeline "your topic" — AUTO_PROCEED: false, human checkpoint: true  # combine options
/proof-checker "paper/" — reviewer: oracle-pro                                # Pro-level proof verification

Important: Codex MCP uses the model from ~/.codex/config.toml, not from skill files. Make sure it says model = "gpt-5.5" (recommended). Other options: gpt-5.3-codex, gpt-5.2-codex, o3. Run codex setup or edit the file directly.

Want Codex to execute but Claude Code to review? See docs/CODEX_CLAUDE_REVIEW_GUIDE.md. That path installs the base skills/skills-codex/*, then overlays skills/skills-codex-claude-review/*, and routes review-heavy skills through the local claude-review MCP bridge.

Want Codex to execute but Gemini to review locally? See docs/CODEX_GEMINI_REVIEW_GUIDE.md and CN. That path installs the base skills/skills-codex/*, then overlays skills/skills-codex-gemini-review/*, and routes the reviewer-aware predefined skills through the local gemini-review MCP bridge using direct Gemini API by default.

Want the Codex mirror install chain? Use tools/install_aris_codex.sh for managed project installs and tools/smart_update_codex.sh for copied Codex installs. The Claude scripts remain the mainline entry points for Claude projects.

See full setup guide for details and alternative model combinations if you don't have Claude/OpenAI API.

🧠 Smart-update workflow — git pull + dry-run + apply; detects personal customizations and only updates safe skills.

cd Auto-claude-code-research-in-sleep
git pull
bash tools/smart_update.sh          # dry-run: shows what's new/changed/safe
bash tools/smart_update.sh --apply  # apply: adds new + updates safe ones

Compares local skills with upstream, detects personal customizations (server paths, API keys, etc.), and only updates skills that are safe to replace. Skills with your personal info are flagged for manual review.

✨ Features

• 📊 74 composable skills — mix and match, or chain into full pipelines (/idea-discovery, /auto-review-loop, /paper-writing, /research-pipeline). See full catalog →

• 🔍 Literature & novelty — multi-source paper search (Zotero + Obsidian + local PDFs + arXiv/Scholar) + cross-model novelty verification

• 💡 Idea discovery — literature survey → brainstorm 8-12 ideas → novelty check → GPU pilot experiments → ranked report

• 🔄 Auto review loop — 4-round autonomous review, 5/10 → 7.5/10 overnight with 20+ GPU experiments

• 📝 Paper writing — narrative → outline → figures → LaTeX → PDF → auto-review (4/10 → 8.5/10), one command. Anti-hallucination citations via DBLP/CrossRef

• 🤖 Cross-model collaboration — Claude Code executes, GPT-5.4 xhigh reviews. Adversarial, not self-play. Optional: — reviewer: oracle-pro → GPT-5.4 Pro via Oracle

• 📝 Peer review — review others' papers as a conference reviewer, with structured scoring and meta-review

• 🖥️ Review-driven experiments — when GPT-5.4 says "run an ablation", Claude auto-writes the script, rsyncs to GPU, runs in screen, collects results, folds back into the paper. Configure server in CLAUDE.md (setup), or rent from Vast.ai with gpu: vast

• 🔀 Flexible models — default Claude × GPT-5.4, also supports GLM, MiniMax, Kimi, LongCat, DeepSeek, etc. — no Claude or OpenAI API required

• 🛑 Human-in-the-loop — configurable checkpoints at key decisions. AUTO_PROCEED=true for full autopilot, false to approve each step

• 📱 Feishu/Lark notifications — three modes: off (default, recommended), push-only (webhook → mobile), interactive (approve/reject in Feishu). Zero impact when off

  <details> <summary>Preview: Push cards (group) &amp; Interactive chat (private)</summary>

  Push Only — group chat cards (experiment done, checkpoint, error, pipeline complete):

  <img src="assets/feishu_push.png" width="700" />

  Interactive — private chat with Claude Code (approve/reject, custom instructions):

  <img src="assets/feishu_interactive.jpg" width="700" /> </details>

• 📚 Research Wiki — persistent knowledge base across papers/ideas/experiments/claims. Failed ideas become anti-repetition memory — ARIS gets smarter every run. Inspired by Karpathy's LLM Wiki

• 🧩 Extensible — domain-specific skills welcome! Add a SKILL.md and open a PR. See community skills like dse-loop (architecture/EDA)

📈 Score Progression (Real Run)

A real overnight 4-round run on an ML research project, from borderline reject to submission-ready:

The loop autonomously ran 20+ GPU experiments, rewrote the paper's narrative framing, and killed claims that didn't hold up — all without human intervention.

🏆 Community Showcase — Papers Built with ARIS

Real projects where the ARIS pipeline was used end-to-end to produce submitted manuscripts. This section does not claim official acceptance unless a row explicitly says so: ratings and quoted verdicts are AI/third-party review signals from tools such as CSPaper and Stanford Agentic Reviewer, not venue decisions. One important caveat: ARIS is designed to optimize through AI-review loops, so elevated AI-review scores are a normal consequence of the workflow rather than independent proof of acceptance. Human reviewers can still bring updated literature knowledge, community context, venue-specific taste, and objections that an AI reviewer did not model. If you've used ARIS to complete a paper, we'd love to feature it here — open an issue or PR!

Papers built with ARIS — from idea to submission. Know more? Let us know!

🧩 Awesome Community Skills & Extensions

Domain-specific skills and external projects contributed by the community. PRs welcome — just add a skills/your-skill/SKILL.md and open a PR!

💡 How to use: Community skills are not auto-wired into core workflows. To use one, ask your executor (Claude Code / OpenClaw / etc.) to read the skill's SKILL.md, then plug it into the appropriate workflow stage based on the description below.

🎉 Community Skills (13): research-refine · experiment-plan · grant-proposal · paper-poster · paper-slides · mermaid-diagram · proof-writer · comm-lit-review · dse-loop · idea-discovery-robot · formula-derivation · paper-illustration · writing-systems-papers

🌐 External Projects & Docs (13): rosetta · open-source-hardening-skills · CitationClaw · auto-hparam-tuning · paper-to-course · deep-research-skills · Antigravity Adaptation Guide · OpenClaw Adaptation Guide · Cursor Adaptation Guide · Codex+Claude Review Bridge · Trae Adaptation Guide · paper-illustration · MiniMax-AI/cli

🙌 Thanks to every contributor! We fold the tables below to keep the README readable — but every skill and project here is equally valued. PRs always welcome!

🔄 Workflows

These skills compose into a full research lifecycle. The four workflows can be used independently or chained together:

• Exploring a new area (e.g., writing a survey)? Start with Workflow 1 → /idea-discovery
• Have a plan, need to implement and run? Workflow 1.5 → /experiment-bridge
• Already have results, need iterative improvement? Workflow 2 → /auto-review-loop
• Ready to write the paper? Workflow 3 → /paper-writing (or step by step: /paper-plan → /paper-figure → /paper-write → /paper-compile → /auto-paper-improvement-loop)
• Got reviews back? Need to rebuttal? Workflow 4 → /rebuttal — parse reviews, draft safe rebuttal, follow-up rounds
• Full pipeline? Workflow 1 → 1.5 → 2 → 3 → submit → 4 → /research-pipeline + /rebuttal — from idea to acceptance
• Want ARIS to remember and learn? 📚 /research-wiki init — persistent memory across sessions. Papers, ideas, failed experiments compound over time
• Want ARIS to improve itself? Workflow M → /meta-optimize — analyze usage logs, propose skill improvements, reviewer-gated

⚠️ Important: These tools accelerate research, but they don't replace your own critical thinking. Always review generated ideas with your domain expertise, question the assumptions, and make the final call yourself. The best research comes from human insight + AI execution, not full autopilot.

Full Pipeline 🚀

/research-lit → /idea-creator → /novelty-check → /research-refine → /experiment-bridge → /auto-review-loop → /paper-writing → submit → /rebuttal → accept! 🎉
  (survey)      (brainstorm)    (verify novel)   (refine method)   (implement+deploy)  (review & fix)      (write paper)   (send)   (reply to reviewers)
  ├────────────── Workflow 1: Idea Discovery ──────────────┤ ├ Workflow 1.5 ─┤ ├── Workflow 2 ──┤ ├── Workflow 3 ──┤         ├── Workflow 4 ──┤

                                     📚 research-wiki (persistent memory — papers, ideas, experiments, claims)
                                        ↕ reads before ideation, writes after every stage, failed ideas = anti-repetition memory

                                              /meta-optimize (Workflow M — runs independently, improves ARIS itself)
                                                 ↑ reads .aris/meta/events.jsonl (accumulated from all runs above)

📝 Blog post: 梦中科研全流程开源

Workflow 1: Idea Discovery & Method Refinement 🔍

"What's the state of the art? Where are the gaps? How do we solve it?"

Don't have a concrete idea yet? Just give a research direction — /idea-discovery handles the rest:

1. 📚 Survey the landscape (recent papers, open problems, recurring limitations)
2. 🧠 Brainstorm 8-12 concrete ideas via GPT-5.4 xhigh
3. 🔍 Filter by feasibility, compute cost, and quick novelty search
4. 🛡️ Validate top ideas with deep novelty check + devil's advocate review
5. 🧪 Pilot top 2-3 ideas in parallel on different GPUs (30 min - 2 hr each)
6. 🏆 Rank by empirical signal — ideas with positive pilot results rise to the top
7. 🔬 Refine the top idea into a problem-anchored proposal via iterative GPT-5.4 review
8. 🧪 Plan claim-driven experiments with ablations, budgets, and run order

The output is a ranked IDEA_REPORT.md plus a refined proposal (refine-logs/FINAL_PROPOSAL.md) and experiment plan (refine-logs/EXPERIMENT_PLAN.md) for the top idea. Dead-end ideas are documented too, saving future exploration.

┌─────────────────────────────────────────────────────────────────┐
│              Idea Discovery & Method Refinement                  │
│                                                                  │
│   /research-lit    /idea-creator    /novelty-check               │
│   (find papers)    (brainstorm)     (verify novelty)             │
│         │               │                │                       │
│         ▼               ▼                ▼                       │
│   ┌──────────┐    ┌──────────┐     ┌──────────┐                │
│   │ Scan     │───▶│ Generate │────▶│ Check if │                │
│   │ local    │    │ 8-12     │     │ idea is  │                │
│   │ papers + │    │ ideas    │     │ novel    │                │
│   │ search   │    │ + rank   │     │          │                │
│   └──────────┘    └──────────┘     └──────────┘                │
│                         │                │                       │
│                         ▼                ▼                       │
│                   ┌──────────┐     ┌──────────┐                │
│                   │ Filter   │────▶│ External │                │
│                   │ by cost, │     │ LLM      │                │
│                   │ novelty  │     │ evaluates│                │
│                   └──────────┘     └──────────┘                │
│                                          │                       │
│                   /research-refine       ▼                       │
│                   (refine method)   ┌──────────┐                │
│                         │          │ Freeze   │                │
│                         ▼          │ problem  │                │
│                   ┌──────────┐     │ anchor + │                │
│                   │ Iterate  │◀───▶│ refine   │                │
│                   │ until    │     │ method   │                │
│                   │ score≥9  │     └──────────┘                │
│                   └──────────┘          │                       │
│                         │               ▼                       │
│                   /experiment-plan  ┌──────────┐                │
│                         │          │ Claim-   │                │
│                         ▼          │ driven   │                │
│                   ┌──────────┐     │ experiment│               │
│                   │ Plan     │────▶│ roadmap  │                │
│                   │ runs     │     └──────────┘                │
│                   └──────────┘                                  │
│                                                                  │
│   Typical flow:                                                  │
│   1. /research-lit "discrete diffusion models"                   │
│   2. /idea-creator "DLLMs post training"                         │
│   3. Review ranked ideas, pick top 2-3                           │
│   4. /novelty-check "top idea" (deep verification)               │
│   5. /research-review "top idea" (critical feedback)             │
│   6. /research-refine "top idea" (problem anchor + method)       │
│   7. /experiment-plan (claim-driven roadmap)                     │
│   8. /run-experiment → /auto-review-loop                         │
└─────────────────────────────────────────────────────────────────┘

Skills involved: research-lit + idea-creator + novelty-check + research-review + research-refine-pipeline

💡 One-command shortcut: /idea-discovery "your research direction" runs this entire workflow automatically.

🔄 Human-in-the-loop: Each phase presents results and waits for your feedback. Not happy? Tell it what's missing — it refines the prompt and regenerates. Trust the defaults? It auto-proceeds with the top-ranked option. You decide how hands-on to be.

⚙️ Pilot experiment budgets (max hours, timeout, GPU budget) are configurable — see Customization.

📝 Blog post: Claude Code 两月 NeurIPS 指北

Workflow 1.5: Experiment Bridge 🔗

"I have a plan. Now implement it, deploy it, and get me initial results."

Already have an experiment plan (from Workflow 1 or your own)? /experiment-bridge turns it into running code:

1. 📋 Parse the experiment plan (refine-logs/EXPERIMENT_PLAN.md)
2. 💻 Implement experiment scripts (reuse existing code, add proper argparse/logging/seeds)
3. 🔍 GPT-5.4 code review — cross-model review catches logic bugs before wasting GPU hours (code review: true by default)
4. ✅ Sanity check — run the smallest experiment first to catch runtime bugs
5. 🚀 Deploy full experiment suite to GPU via /run-experiment
6. 📊 Collect initial results and update the experiment tracker

┌─────────────────────────────────────────────────────────────────┐
│                Workflow 1.5: Experiment Bridge                    │
│                                                                  │
│   EXPERIMENT_PLAN.md                                             │
│         │                                                        │
│         ▼                                                        │
│   ┌──────────┐     ┌──────────┐     ┌──────────┐               │
│   │ Claude   │────▶│ GPT-5.4  │────▶│ Sanity   │               │
│   │ Code     │     │ xhigh    │     │ Check    │               │
│   │ writes   │     │ reviews  │     │ (1 GPU)  │               │
│   │ code     │     │ code     │     │          │               │
│   └──────────┘     └──────────┘     └──────────┘               │
│                                          │                       │
│                                          ▼                       │
│   ┌──────────┐     ┌──────────┐     ┌──────────┐               │
│   │ Collect  │◀────│ Monitor  │◀────│ Deploy   │               │
│   │ results  │     │ progress │     │ to GPUs  │               │
│   │          │     │ (+ W&B)  │     │          │               │
│   └──────────┘     └──────────┘     └──────────┘               │
│         │                                                        │
│         ▼                                                        │
│   Ready for /auto-review-loop                                    │
└─────────────────────────────────────────────────────────────────┘

Skills involved: experiment-bridge + run-experiment + monitor-experiment

💡 One-command shortcut: /experiment-bridge reads refine-logs/EXPERIMENT_PLAN.md automatically. Or point it to any plan: /experiment-bridge "my_plan.md".

⚙️ CODE_REVIEW, AUTO_DEPLOY, SANITY_FIRST, MAX_PARALLEL_RUNS are configurable — see Customization.

Workflow 2: Auto Research Loop 🔁 (sleep & wake up to results)

"Review my paper, fix what's wrong, repeat until it's good."

GPT-5.4 reviews → identifies weaknesses → suggests experiments → Claude Code writes scripts, deploys to GPU, monitors results, rewrites the paper — all while you sleep. Just add your GPU server config to CLAUDE.md.

1. 🔍 Deep review — GPT-5.4 xhigh reviews the current paper / claims / experiments and identifies weaknesses
2. 🩹 Fix — Claude implements the fixes (rewrites sections, adds baselines, or runs new experiments via /run-experiment); skips any experiment estimated > 4 GPU-hours and flags it for manual follow-up
3. 📊 Re-evaluate — collect results via /monitor-experiment, update paper, feed back to the reviewer
4. 🔁 Repeat — until score ≥ POSITIVE_THRESHOLD (default 6/10) or MAX_ROUNDS (default 4) is hit; if context window fills mid-loop, the workflow auto-resumes from REVIEW_STATE.json

┌─────────────────────────────────────────────────────────────┐
│                    Auto Review Loop                          │
│                                                              │
│   /research-review          /auto-review-loop                │
│   (single deep review)      (autonomous loop)                │
│         │                         │                          │
│         ▼                         ▼                          │
│   ┌──────────┐   ┌──────────┐   ┌──────────┐               │
│   │ External  │──▶│ Implement│──▶│ Monitor  │──▶ repeat     │
│   │ LLM      │   │ fixes    │   │ results  │    until       │
│   │ reviews  │   │ & run    │   │          │    score ≥ 6   │
│   └──────────┘   │ experiments│  └──────────┘               │
│                   └──────────┘                               │
│                                                              │
│   When reviewer suggests a new method direction:             │
│   /novelty-check — verify idea isn't already published       │
│                                                              │
│   Supporting skills:                                         │
│   /run-experiment    — deploy to local/remote/vast.ai GPU     │
│   /analyze-results   — interpret experiment outputs          │
│   /monitor-experiment — check progress, collect results      │
└─────────────────────────────────────────────────────────────┘

Skills involved: auto-review-loop + research-review + novelty-check + run-experiment + analyze-results + monitor-experiment

💡 One-command shortcut: /auto-review-loop "your paper topic" runs this entire workflow automatically.

What to pass as argument? A short topic or scope is enough — the skill automatically reads your project's narrative docs (NARRATIVE_REPORT.md), memory files, experiment results, and prior reviews to build the full context for GPT-5.4. Examples:

• /auto-review-loop "factorized gap in discrete diffusion LMs" — broad topic, skill finds everything
• /auto-review-loop "focus on Section 3-5, our CRF results are weak" — targeted scope with hints
• /auto-review-loop — also works: skill reads project files and infers the topic

🎮 Reviewer Difficulty — control how adversarial the reviewer is:

/auto-review-loop "topic" — difficulty: nightmare    # GPT reads your code and verifies claims itself

🛡️ Key safety features:

• 🔒 MAX_ROUNDS = 4 — prevents infinite loops; stops early if score threshold is met
• ⏱️ > 4 GPU-hour experiments skipped — won't launch massive jobs; flags them for manual follow-up
• 🧠 Prefer reframing over new experiments — when both can address a weakness, chooses the cheaper path
• 🪞 No hiding weaknesses — explicit rule: "Do NOT hide weaknesses to game a positive score"
• 🔧 Fix before re-review — must actually implement fixes before resubmitting; no empty promises
• 💾 Compact recovery — persists state (REVIEW_STATE.json) after each round. If the context window fills up and auto-compacts mid-loop, the workflow reads the state file and resumes from where it left off — no human intervention needed

⚙️ MAX_ROUNDS, score threshold, and GPU limits are configurable — see Customization.

📝 Blog post: 开源 | 睡觉 Claude 自动跑实验改文

Workflow 3: Paper Writing Pipeline 📝

"Turn my research narrative into a submission-ready PDF." Requires a local LaTeX environment — see Prerequisites.

1. 📝 Narrate — write NARRATIVE_REPORT.md (claims, experiments, results, figure descriptions); see templates/NARRATIVE_REPORT_TEMPLATE.md
2. 🧭 Plan — /paper-plan builds the claims-evidence matrix + section plan
3. 📊 Figures — /paper-figure generates data-driven plots and comparison tables from JSON/CSV
4. ✍️ Write — /paper-write produces section-by-section LaTeX
5. 🔧 Compile — /paper-compile builds the PDF, fixes errors, runs the page-limit check
6. ✨ Improve — /auto-paper-improvement-loop runs 2 rounds of GPT-5.4 content review + final format check

┌─────────────────────────────────────────────────────────────┐
│                   Paper Writing Pipeline                      │
│                                                               │
│   /paper-plan      /paper-figure     /paper-write             │
│   (outline)        (plots & tables)  (LaTeX draft)            │
│        │                │                 │                   │
│        ▼                ▼                 ▼                   │
│   ┌──────────┐    ┌──────────┐     ┌──────────┐              │
│   │ Claims-  │───▶│ Generate │────▶│ Section  │──┐           │
│   │ Evidence │    │ figures, │     │ by       │  │           │
│   │ Matrix + │    │ tables,  │     │ section  │  │           │
│   │ Section  │    │ LaTeX    │     │ LaTeX    │  │           │
│   │ Plan     │    │ includes │     │ draft    │  │           │
│   └──────────┘    └──────────┘     └──────────┘  │           │
│        │                                          │           │
│        │         /paper-compile                   │           │
│        │         (build PDF)                      │           │
│        │              │                           │           │
│        ▼              ▼                           ▼           │
│   ┌──────────────────────────────────────────────────┐       │
│   │ NARRATIVE_REPORT.md ──► PAPER_PLAN.md ──► paper/ │       │
│   │    (input)             (outline)      (LaTeX+PDF)│       │
│   └──────────────────────────────────────────────────┘       │
│                                                               │
│   Typical flow:                                               │
│   1. Write NARRATIVE_REPORT.md (from Workflow 2 results)      │
│   2. /paper-plan (claims-evidence matrix + section plan)      │
│   3. /paper-figure (comparison tables, training curves, etc.) │
│   4. /paper-write (section-by-section LaTeX generation)       │
│   5. /paper-compile (build PDF, fix errors, page check)       │
│   6. /auto-paper-improvement-loop (review ×2 + format check)  │
└─────────────────────────────────────────────────────────────┘

Skills involved: paper-plan + paper-figure + paper-write + paper-compile + auto-paper-improvement-loop + (post-acceptance) paper-poster + paper-slides

One-command shortcut: /paper-writing "NARRATIVE_REPORT.md" runs this entire workflow automatically.

Input: A NARRATIVE_REPORT.md describing the research: claims, experiments, results, figures. The more detailed the narrative (especially figure descriptions and quantitative results), the better the output.

Output: A paper/ directory with LaTeX source, clean .bib (only cited entries), and compiled PDF. The PDF is labelled submission-ready only when run at — effort: max | beast (or explicit — assurance: submission) and tools/verify_paper_audits.sh reports green on the three mandatory audits (proof-checker, paper-claim-audit, citation-audit); see Assurance Gate below. At the default balanced level, the output is a reviewed draft.

Key features:

• 📐 Claims-Evidence Matrix — every claim maps to evidence, every experiment supports a claim
• 📊 Auto figure generation — line plots, bar charts, comparison tables from JSON data
• 🧹 Clean bib — automated filtering removes uncited entries (948→215 lines in testing). Real BibTeX from DBLP/CrossRef instead of LLM-generated entries
• 📄 Flexible sections — 5-8 sections depending on paper type (theory papers often need 7)
• 🔍 GPT-5.4 review — each step optionally reviewed by external LLM
• ✂️ De-AI polish — removes AI writing patterns (delve, pivotal, landscape...)
• 🎯 Page verification — pdftotext-based precise check that main body fits page limit

⚠️ Figure generation scope: /paper-figure auto-generates data-driven plots (training curves, bar charts, heatmaps) and comparison tables from JSON/CSV. For architecture diagrams and method figures: illustration: gemini (default) uses Claude→Gemini→Nano Banana Pro for publication-quality diagrams; illustration: mermaid generates Mermaid diagrams for free; illustration: false skips AI figures entirely.

Gemini API setup (for illustration: gemini): Get your API key at Google AI Studio, then set it as an environment variable: export GEMINI_API_KEY="your-key". Or add to your shell profile (~/.zshrc / ~/.bashrc). No other dependencies needed.

Tested end-to-end: Generated a 9-page ICLR 2026 theory paper (7 sections, 29 citations, 4 figures, 2 comparison tables) from a single NARRATIVE_REPORT.md — zero compilation errors, zero undefined references.

Auto Paper Improvement Loop ✨

After Workflow 3 generates the paper, /auto-paper-improvement-loop runs 2 rounds of GPT-5.4 xhigh content review → fix → recompile, plus a final format compliance check, autonomously polishing the paper from rough draft to a reviewer-scored draft. Whether the result is tagged submission-ready is decided separately by the Phase 6 assurance gate (see Assurance Gate).

Score Progression (Real Test — ICLR 2026 theory paper):

Final: 8 pages main body (ICLR limit: 9), 0 overfull hbox, ICLR-compliant. +4.5 points across 3 rounds.

1. CRITICAL — Assumption-model mismatch: A boundedness assumption contradicted the model's distributional family. Replaced with a tail-compatible assumption and added formal truncation bridge.
2. CRITICAL — Theory-practice gap: Theory assumes idealized encoders, experiments use learned nonlinear encoders. Softened "validate" → "demonstrate practical relevance" and added explicit disclaimer.
3. MAJOR — Missing quantitative metrics: Added parameter count table (latent vs total) with honest accounting of system cost.
4. MAJOR — Theorem not self-contained: Added "Interpretation" paragraph listing all dependencies explicitly.
5. MAJOR — Overclaim in novelty statement: Scoped a broad "first convergence guarantee" to precise conditions under which it holds.
6. MAJOR — Notation confusion: Renamed a symbol that clashed with another key variable. Added Notation paragraph.

1. MAJOR — Missing theory-aligned experiments: Added a synthetic validation subsection directly testing the two main theoretical predictions under controlled conditions.
2. MAJOR — Overclaim softening: Replaced strong equivalence claims with appropriately hedged language across all files.
3. MAJOR — Informal theoretical argument: Formalized an informal justification into a proper proposition with explicit error bounds.
4. MINOR — Weak limitations: Expanded to explicitly list all assumptions and acknowledge missing standard evaluations.

1. Removed hero figure block (saved ~0.7 pages)
2. Compressed conclusion from 15→9 lines
3. Moved synthetic validation to Appendix A
4. Moved comparison tables to Appendix B
5. Fixed overfull hbox (85pt) with \resizebox
6. Added compact float spacing (\captionsetup, \textfloatsep)
7. Inlined centered question block in introduction
8. Tightened itemize environments

Workflow 4: Rebuttal 📝 (reply to reviewers safely)

"Reviews are in. Help me draft a safe, grounded rebuttal."

Got reviews back? /rebuttal parses them, builds a strategy, and drafts a venue-compliant response:

1. 📋 Parse — normalize reviews, validate venue rules (character limit, text-only, etc.)
2. 🔍 Atomize — split each review into issue cards (type, severity, reviewer stance)
3. 🗺️ Strategize — global themes, per-reviewer priorities, character budget, blocked claims
4. 🧪 Evidence sprint — if auto experiment: true, auto-run supplementary experiments via /experiment-bridge
5. ✍️ Draft — global opener + numbered per-reviewer responses + closing for meta-reviewer
6. 🛡️ Safety check — 6 lints: coverage, provenance, commitment, tone, consistency, limit
7. 🔬 GPT-5.4 stress test — internal skeptical review of the draft
8. 📄 Finalize — two outputs: PASTE_READY.txt (exact character count) + REBUTTAL_DRAFT_rich.md (extended version for manual editing)
9. 🔄 Follow-up rounds — delta replies for reviewer discussions, technically escalating

┌─────────────────────────────────────────────────────────────────┐
│                   Workflow 4: Rebuttal                            │
│                                                                  │
│   Reviews arrive                                                 │
│         │                                                        │
│         ▼                                                        │
│   ┌──────────┐     ┌──────────┐     ┌──────────┐               │
│   │ Parse &  │────▶│ Strategy │────▶│ Evidence  │               │
│   │ atomize  │     │ plan     │     │ sprint    │               │
│   │ reviews  │     │          │     │ (optional)│               │
│   └──────────┘     └──────────┘     └──────────┘               │
│                                          │                       │
│                                          ▼                       │
│   ┌──────────┐     ┌──────────┐     ┌──────────┐               │
│   │ Finalize │◀────│ GPT-5.4  │◀────│ Draft    │               │
│   │ 2 versions│    │ stress   │     │ rebuttal │               │
│   │          │     │ test     │     │          │               │
│   └──────────┘     └──────────┘     └──────────┘               │
│         │                                                        │
│         ▼                                                        │
│   PASTE_READY.txt (strict) + RICH.md (extended)                  │
│         │                                                        │
│         ▼                                                        │
│   Follow-up rounds (delta replies, per-reviewer threads)         │
└─────────────────────────────────────────────────────────────────┘

Skills involved: rebuttal

💡 Quick mode: /rebuttal — quick mode: true stops after parsing + strategy (Phase 0-3). See what reviewers want before committing to a full draft.

⚙️ VENUE, AUTO_EXPERIMENT, QUICK_MODE, MAX_STRESS_TEST_ROUNDS are configurable — see Customization.

Three safety gates — rebuttal will NOT finalize if any fails:

• 🔒 Provenance — every claim maps to paper/review/user-confirmed result. No fabrication.
• 🔒 Commitment — every promise is user-approved. No overpromising.
• 🔒 Coverage — every reviewer concern is tracked. Nothing disappears.

Workflow 5: Resubmit Pipeline 🔁 (port a paper to a new venue, text-only)

"Paper accepted somewhere or rejected from venue A. Port it to venue B under hard constraints."

/resubmit-pipeline ports a polished paper from one venue to another with strict guardrails — no new experiments, no bib edits, no framework changes, never overwrites prior submissions. Use it for journal version of a conference paper, ML venue → other ML venue, anonymized re-submit after a non-anonymous workshop. Not for major revisions (use /paper-writing for those).

1. 📁 Physical isolation — copy into <NEW_VENUE_DIR>/; the original submission directory is never touched.
2. 🛡️ 5-layer anonymity check — author names, affiliations, self-citations, GitHub / Overleaf URLs, in-text "we" patterns that break double-blind.
3. 🔬 Audits (soft-only) — /proof-checker, /paper-claim-audit, /citation-audit --soft-only. The --soft-only mode translates KEEP/FIX/REPLACE/REMOVE verdicts to text-rewrite proposals when the bib is frozen; hallucinated citations get a drop_cite_in_body_only action.
4. ✏️ Microedit — /auto-paper-improvement-loop --edit-whitelist <path> with a YAML schema (allowed_paths / forbidden_paths / forbidden_operations like new_cite / new_theorem_env / numerical_claim, forbidden_deletions, max_edits_per_round) + per-round diff gate.
5. 🗡 Adversarial gate — /kill-argument final attack/adjudication pass; rejection if any still_unresolved at critical severity.
6. 📤 Compile + push — /paper-compile + optional /overleaf-sync push.

┌──────────────────────────────────────────────────────────────────────┐
│              Workflow 5: Text-only Resubmit                          │
│                                                                      │
│  Polished paper                                                      │
│       │                                                              │
│       ▼                                                              │
│  Isolate → Anonymity (5-layer) → Audits (--soft-only)                │
│       │                                                              │
│       ▼                                                              │
│  Microedit (whitelist + diff gate) → /kill-argument adversarial gate │
│       │                                                              │
│       ▼                                                              │
│  Compile + Overleaf push     →    <NEW_VENUE_DIR>/                   │
└──────────────────────────────────────────────────────────────────────┘

Skills involved: resubmit-pipeline (orchestrator), auto-paper-improvement-loop --edit-whitelist, citation-audit --soft-only, proof-checker, paper-claim-audit, kill-argument, paper-compile, overleaf-sync (optional)

Hard constraints (cannot be overridden):

• 🔒 No new experiments — every numerical claim must already exist in the source paper.
• 🔒 No bib edits — citation issues become body-text rewrites via --soft-only.
• 🔒 No framework changes — theorem environment, claim shape, contribution scope are frozen.
• 🔒 Never overwrites prior submissions — the new venue gets its own directory.

Master ledger: RESUBMIT_REPORT.json with the 7-verdict failure-mode table (including USER_DECISION runtime state) per shared-references/assurance-contract.md. See the 2026-05-05 News entry for the full feature breakdown.

Workflow 6: Conference Talk Pipeline 🎤 (paper → slides → polish → audits)

"Paper is in. Now prepare the conference talk."

/paper-talk orchestrates the full talk-prep flow as a sister workflow to /paper-writing and /paper-poster. /slides-polish is the post-generation visual pass invoked internally — you do not need to call it separately.

1. 📋 Outline — extract from paper/ (or NARRATIVE_REPORT.md); one slide-cluster per contribution; map sections to talk beats.
2. 🎨 Generate — /paper-slides produces Beamer source + PPTX + speaker notes + Q&A prep.
3. 💎 Polish — /slides-polish per-page Codex review against the reference PDF, applying a fix-pattern catalog (PPTX font scaling 1.5-1.8× for projector legibility, text-frame resize after font bump, banner-as-tcolorbox, italic style leak guard, em-dash spacing, Chinese EA font hint via PingFang SC, anonymity placeholder discipline).
4. 🛡️ Audit (when assurance: conference-ready) — /paper-claim-audit + /citation-audit run against a synthetic paper directory at .aris/paper-talk/audit-input/sections/*.tex + symlinked .bib / results/ / figures/. Each emits a 6-state JSON verdict per shared-references/assurance-contract.md; non-green blocks the Final Report.

┌──────────────────────────────────────────────────────────────────────┐
│             Workflow 6: Conference Talk                              │
│                                                                      │
│  paper/  →  outline  →  /paper-slides  (Beamer + PPTX + notes)       │
│                                  │                                   │
│                                  ▼                                   │
│                         /slides-polish  (per-page Codex pass)        │
│                                  │                                   │
│                                  ▼                                   │
│               assurance: conference-ready ?                          │
│                 ├─ yes → /paper-claim-audit + /citation-audit        │
│                 │        on synthetic-paper staging adapter          │
│                 │        → 6-state verdict gates Final Report        │
│                 └─ no  → Final Report directly                       │
└──────────────────────────────────────────────────────────────────────┘

Skills involved: paper-talk (orchestrator), paper-slides, slides-polish, paper-claim-audit + citation-audit (at assurance: conference-ready)

Assurance ladder (independent from the effort axis): draft / polished (default) / conference-ready. Legal combination: — effort: lite, assurance: conference-ready means "fast pipeline, every audit must emit a verdict before final report."

Standalone slide / poster tools: if you only want the artifact and not the full orchestration, /paper-slides "paper/" and /paper-poster "paper/" work directly without /paper-talk. See the 2026-05-06 News entry for the full feature breakdown.

📚 Research Wiki — Persistent Research Memory

"Stop re-deriving. Start compounding." — inspired by Karpathy's LLM Wiki

Without the wiki, ARIS is stateless — every /idea-discovery starts from scratch. With the wiki, ARIS accumulates knowledge across the entire research lifecycle: papers read, ideas tested, experiments run, claims verified or invalidated.

The key insight: failed ideas are the most valuable memory. A researcher who knows what doesn't work generates better ideas than one starting from zero.

Setup:

> /research-wiki init     # one-time, creates research-wiki/ in your project

That's it. Once initialized, the wiki works automatically:

Four entity types:

Typed relationships (stored in graph/edges.jsonl):

paper --extends--> paper              idea --inspired_by--> paper
paper --contradicts--> paper          idea --tested_by--> experiment
paper --addresses_gap--> gap          experiment --supports--> claim
paper --supersedes--> paper           experiment --invalidates--> claim

Spiral learning in action:

Round 1: read 15 papers → wiki remembers → idea A → experiment → FAIL
         wiki records: "A fails because OOM at batch>32, loss diverges"

Round 2: /idea-creator reads wiki → sees A failed → generates idea D (avoids A's trap)
         → experiment → PARTIAL SUCCESS
         wiki records: "D works on small models, fails on large"

Round 3: /idea-creator reads wiki → knows A failed + D partial → generates idea F
         (combines D's success with new approach) → experiment → SUCCESS 🎉

Subcommands:

/research-wiki init                              # initialize wiki
/research-wiki ingest "paper title" — arxiv: xxx  # manually add a paper
/research-wiki query "topic"                      # rebuild query_pack.md
/research-wiki update idea:001 — outcome: negative # update entity
/research-wiki lint                               # health check (orphans, contradictions, stale claims)
/research-wiki stats                              # overview (paper/idea/experiment/claim counts)

🔒 Safe by design: All workflow hooks are guarded by if research-wiki/ exists. No wiki = no impact. Zero dependencies (pure Python stdlib). You choose when to enable it.

Workflow M: Meta-Optimize 🧬 (ARIS optimizes itself)

"Analyze my usage patterns and improve your own skills."

Unlike Workflows 1–4 which optimize research artifacts (papers, code, experiments), Workflow M optimizes the harness itself — the SKILL.md instructions, default parameters, and convergence rules that govern how ARIS operates. Inspired by Meta-Harness (Lee et al., 2026).

Setup (one-time, in normal terminal):

mkdir -p .claude .aris/meta tools/meta_opt
cp Auto-claude-code-research-in-sleep/templates/claude-hooks/meta_logging.json .claude/settings.json
cp Auto-claude-code-research-in-sleep/tools/meta_opt/*.sh tools/meta_opt/
chmod +x tools/meta_opt/*.sh
claude   # hooks active immediately

Usage (after 5+ workflow runs):

> /meta-optimize                        # analyze current project
> /meta-optimize "auto-review-loop"     # focus on one skill
> /meta-optimize --global               # analyze trends across ALL projects
> /meta-optimize apply 1                # apply recommended change #1

How it works:

1. 📊 Passive logging — Claude Code hooks silently record every skill invocation, tool call, failure, parameter override, and user prompt. Events are written to both project-level (.aris/meta/events.jsonl) and global (~/.aris/meta/events.jsonl, with a "project" tag) logs. Zero user effort.
2. 🔍 Pattern analysis — /meta-optimize reads the log and identifies:
   • Parameters users override most often (bad defaults)
   • Tools that fail repeatedly in specific skills (missing error handling)
   • Review score plateaus (convergence rules too loose/tight)
   • Manual corrections users make (skill gaps)
3. 🩹 Patch proposal — generates minimal diffs to target SKILL.md files with data-backed justifications
4. 🔬 Reviewer gate — GPT-5.4 xhigh reviews each patch: does the evidence support it? could it hurt other users?
5. ✅ User approval — only applied with explicit user consent. All changes are logged and reversible.

┌─────────────────────────────────────────────────────────────────┐
│                  Workflow M: Meta-Optimize                        │
│                                                                  │
│   Normal ARIS usage (W1-W4)                                      │
│         │ (hooks log events passively)                           │
│         ▼                                                        │
│   .aris/meta/events.jsonl                                        │
│         │                                                        │
│         ▼                                                        │
│   ┌──────────┐     ┌──────────┐     ┌──────────┐               │
│   │ Analyze  │────▶│ Propose  │────▶│ GPT-5.4  │               │
│   │ patterns │     │ SKILL.md │     │ reviews  │               │
│   │          │     │ patches  │     │ patch    │               │
│   └──────────┘     └──────────┘     └──────────┘               │
│                                          │                       │
│                                          ▼                       │
│                                    User approves?                 │
│                                     Yes → Apply                  │
│                                     No  → Skip                   │
└─────────────────────────────────────────────────────────────────┘

What gets optimized (harness components):

What does NOT get optimized: research artifacts (papers, code, experiments) — that's what W1–W4 do.

Skills involved: meta-optimize

💡 This is a maintenance workflow, not part of the W1→W1.5→W2→W3→W4 research pipeline. Run it periodically, like git gc for your research harness.

⚡ Effort Levels

"How hard should ARIS work?" — Every skill accepts — effort: lite | balanced | max | beast.

What NEVER changes regardless of effort:

• Codex reasoning: always xhigh (reviewer quality is non-negotiable)
• DBLP/CrossRef citations: always on
• Reviewer independence: always on
• Experiment integrity: always on

# Every skill accepts effort independently
/research-lit "topic" — effort: beast              # 40-50 papers, 15+ queries
/idea-creator "direction" — effort: lite           # 4-6 ideas, quick filter
/auto-review-loop — effort: max                    # 6 rounds, 4-6 fixes/round

# Mix with specific overrides
/auto-review-loop — effort: beast, review_rounds: 3  # beast everything, but cap at 3 rounds

# Full pipeline
/research-pipeline "your topic" — effort: beast    # top-venue sprint mode

📖 Full specification: shared-references/effort-contract.md

Assurance Gate (effort: max | beast)

ARIS has two independent axes: effort controls how much work is done (breadth/depth), assurance controls whether mandatory audits are load-bearing. Default mapping:

What this fixes: previously, — effort: beast did not actually guarantee the three mandatory audits ran — the content detectors could silent-skip, so beast-mode papers could ship without proof verification or citation checks. The assurance axis makes audit enforcement externally verifiable via tools/verify_paper_audits.sh (the verifier's exit code is the source of truth, not the executor's self-report).

Backwards compatibility: users on the default balanced level see zero change. Only users who opt up to max / beast, or who explicitly pass — assurance: submission, see the new gate.

Escape hatch: — effort: beast, assurance: draft gets the old "depth-only, no audit gate" behavior back. Legal but discouraged for actual submissions.

Optional harness hardening (advanced): teams who want the model to be physically prevented from ending a session while the verifier is red can register a Stop hook in ~/.claude/settings.json (replace <ARIS_REPO> with the absolute path to your ARIS clone, e.g. /Users/you/Auto-claude-code-research-in-sleep):

{
  "hooks": {
    "Stop": [
      {"command": "bash <ARIS_REPO>/tools/verify_paper_audits.sh paper/ --assurance submission"}
    ]
  }
}

This is not required — the default repo behavior (Phase 6 verifier-as-truth) already blocks Final Report emission on a red verdict. The Stop hook is a belt-and-suspenders layer for teams that want harness-level enforcement.

📖 Full specification: shared-references/assurance-contract.md

🧿 Optional: GPT-5.4 Pro via Oracle

For expert researchers who want the strongest possible reviewer.

Oracle unlocks GPT-5.4 Pro as an ARIS reviewer — the strongest reasoning model available. Pro excels at deep mathematical proof verification, line-by-line code auditing, and complex experimental design critique.

Usage rule: add — reviewer: oracle-pro to any reviewer-aware skill (/research-review, /proof-checker, /experiment-audit, /auto-review-loop, /idea-creator, /rebuttal, …).

Default is always Codex xhigh. Oracle not installed = zero impact. — reviewer: oracle-pro without Oracle installed = graceful fallback to Codex + warning.

Setup:

# 1. Install Oracle
npm install -g @steipete/oracle

# 2. Add Oracle MCP to Claude Code
claude mcp add oracle -s user -- oracle-mcp

# 3. Restart Claude Code

# 4a. API mode (fast, recommended):
export OPENAI_API_KEY="your-key"

# 4b. Browser mode (free, no API key — log in to ChatGPT in Chrome):
# Just open Chrome → chatgpt.com → log in

Examples — add — reviewer: oracle-pro to any skill:

/research-review "my draft" — reviewer: oracle-pro          # Pro-level paper critique
/proof-checker "paper/" — reviewer: oracle-pro              # deepest mathematical verification
/experiment-audit — reviewer: oracle-pro                    # Pro audits your eval code
/auto-review-loop "scope" — reviewer: oracle-pro            # Pro stress test each round
/idea-creator "direction" — reviewer: oracle-pro            # Pro evaluates your ideas
/rebuttal "paper/ + reviews" — reviewer: oracle-pro         # Pro stress tests your rebuttal

📖 Full specification: shared-references/reviewer-routing.md

🧰 Skills Catalog

ARIS ships 74+ skills across literature, ideation, experiments, audit, writing, talks, patents, and meta-utilities. The full catalog (with role, category, and requirements per skill) lives in docs/SKILLS_CATALOG.md to keep this README scannable.

Start here for common entry points:

→ Browse all 74 skills by category in the full catalog →

⚙️ Setup

Prerequisites

1. Claude Code installed
2. (For review skills) Codex CLI installed and configured as MCP server:

   npm install -g @openai/codex
   claude mcp add codex -s user -- codex mcp-server
   

3. (For Workflow 3: paper writing) LaTeX environment with latexmk and pdfinfo:

   # macOS
   brew install --cask mactex    # or: brew install basictex
   brew install poppler          # provides pdfinfo
   
   # Ubuntu/Debian
   sudo apt install texlive-full latexmk poppler-utils
   
   # Verify
   latexmk --version && pdfinfo -v
   

   If you only need Workflow 1 & 2 (idea discovery + auto review), LaTeX is not required.

Install Skills

💡 Recommended: project-local flat symlink install (since 2026-04-20). Each ARIS skill is symlinked individually into .claude/skills/<skill-name>, so Claude Code's slash-command discovery picks them up. A manifest at .aris/installed-skills.txt tracks what ARIS installed — uninstall and reconcile only ever touch managed entries, never your own skills.

🤖 Codex mirror route: keep Claude on install_aris.sh / smart_update.sh. For Codex-native project installs, use install_aris_codex.sh; for copied Codex installs, use smart_update_codex.sh.

# 1. Clone ARIS once to a stable location
git clone https://github.com/wanshuiyin/Auto-claude-code-research-in-sleep.git ~/aris_repo

# 2. For each project that uses ARIS, attach via symlinks:
cd ~/your-paper-project
bash ~/aris_repo/tools/install_aris.sh
# → creates one symlink per skill: .claude/skills/<skill> → ~/aris_repo/skills/<skill>
# → writes manifest .aris/installed-skills.txt (tracks every entry ARIS installed)
# → updates managed CLAUDE.md ARIS block (best-effort, compare-and-swap)
# → re-runnable: rerun anytime to reconcile new/removed upstream skills

# 3. To update existing skills' content for ALL attached projects:
cd ~/aris_repo && git pull   # symlinks resolve to live upstream — content updates automatically

# 3a. To pick up newly added or removed upstream skills, rerun the installer:
bash ~/aris_repo/tools/install_aris.sh ~/your-paper-project   # adds new symlinks, removes broken ones

# Other useful flags:
bash ~/aris_repo/tools/install_aris.sh --dry-run        # show plan, no changes
bash ~/aris_repo/tools/install_aris.sh --uninstall      # remove only managed symlinks (per manifest)
bash ~/aris_repo/tools/install_aris.sh --from-old       # migrate from old nested .claude/skills/aris/

# Windows (PowerShell, requires admin or developer mode for junctions):
.\tools\install_aris.ps1 C:\path\to\your-paper-project

Why "git pull" alone isn't enough for new/removed skills: the flat layout uses one symlink per skill, so upstream additions/deletions don't propagate until the installer is re-run. The trade-off bought us Claude Code's automatic slash-command discovery (which only scans one directory level deep).

If you previously installed via install_aris.sh (which created .claude/skills/aris/ as a single nested symlink) or via smart_update.sh --target-subdir .claude/skills/aris, your slash commands probably weren't being auto-discovered by Claude Code. Migrate to the flat layout:

# Symlink-style legacy install:
bash ~/aris_repo/tools/install_aris.sh ~/your-project --from-old

# Copy-style legacy install (with possible local edits — chose strategy explicitly):
bash ~/aris_repo/tools/install_aris.sh ~/your-project --from-old --migrate-copy keep-user
#   → keeps your nested .claude/skills/aris/ copy intact alongside the new flat install
bash ~/aris_repo/tools/install_aris.sh ~/your-project --from-old --migrate-copy prefer-upstream
#   → archives nested copy to .aris/legacy-copy-backup-<timestamp>/, then flattens

Project-local copy (no symlinks, useful for per-project skill edits):

mkdir -p ~/your-project/.claude/skills
bash ~/aris_repo/tools/smart_update.sh --project ~/your-project --apply
# Default --target-subdir is .claude/skills (flat), which is what Claude Code expects.
# (The old --target-subdir .claude/skills/aris is now deprecated — see migration block above.)

Global install (one copy in your home dir, available to every project):

mkdir -p ~/.claude/skills
cp -r ~/aris_repo/skills/* ~/.claude/skills/
# Update with: bash tools/smart_update.sh --apply

Global install increases the risk of skill name collisions with other globally-installed packs. Use only if you don't mix ARIS with Superpowers / OpenHands / etc. — otherwise prefer the project-local install above.

💡 New Claude Code versions may not auto-create ~/.claude/skills/. If using global install, create it first: mkdir -p ~/.claude/skills/. The symlink installer handles directory creation automatically.

codex-plugin-cc provides additional Codex capabilities that ARIS auto-detects when installed:

# In Claude Code:
/plugin marketplace add openai/codex-plugin-cc
/plugin install codex@openai-codex
/reload-plugins
/codex:setup

Where ARIS uses the plugin:

All plugin features are optional — if not installed, ARIS falls back to Claude's own diagnosis. The plugin just adds a second pair of eyes.

Note: ARIS's core cross-model review (paper scoring, idea evaluation, rebuttal stress test) still uses Codex MCP, which allows custom prompts. The plugin cannot replace this.

Update Skills

cd Auto-claude-code-research-in-sleep
git pull

# 🧠 Smart update (recommended) — analyzes what's safe to update
bash tools/smart_update.sh          # dry-run: shows what would change
bash tools/smart_update.sh --apply  # apply: adds new + updates safe ones

# Manual options (if you prefer):
# cp -r skills/* ~/.claude/skills/       # Option A: overwrite all
# cp -rn skills/* ~/.claude/skills/      # Option B: only add new, keep yours
# cp -r skills/experiment-bridge ~/.claude/skills/  # Option C: specific skill

💡 Smart update compares your local skills with upstream, detects personal customizations (server paths, API keys, etc.), and only updates skills that are safe to replace. Skills with your personal info are flagged for manual review.

Usage

# Workflow 1: Idea Discovery
> /idea-discovery "your research direction"          # full pipeline
> /research-lit "topic"                              # just literature survey (all sources)
> /research-lit "topic" — sources: zotero, web        # mix and match sources
> /research-lit "topic" — sources: deepxiv            # DeepXiv-only progressive retrieval
> /research-lit "topic" — sources: exa                # Exa AI-powered web search with content extraction
> /research-lit "topic" — arxiv download: true         # also download top arXiv PDFs
> /arxiv "discrete diffusion" — download               # standalone arXiv search + download
> /idea-creator "topic"                              # just brainstorm

# Workflow 2: Auto Research Loop
> /auto-review-loop "your paper topic"               # review → fix → repeat
> /research-review "your paper"                      # single deep review

# Workflow 3: Paper Writing
> /paper-writing "NARRATIVE_REPORT.md"               # full pipeline
> /paper-plan "NARRATIVE_REPORT.md"                  # just outline
> /paper-compile "paper/"                            # just compile

# Full Pipeline
> /research-pipeline "your research direction"       # Workflow 1 → 2 → 3 end-to-end

# Supporting Skills
> /run-experiment train.py --lr 1e-4 --epochs 100
> /analyze-results figures/*.json
> /monitor-experiment server5

🌙 Auto-Allow for Overnight Runs (Optional)

To run the auto-review loop without clicking permission prompts, add to .claude/settings.local.json:

{
  "permissions": {
    "allow": [
      "mcp__codex__codex",
      "mcp__codex__codex-reply",
      "Write",
      "Edit",
      "Skill(auto-review-loop)"
    ]
  }
}

When GPT-5.4 says "run an ablation study" or "add a baseline comparison", Claude Code automatically writes the experiment script and deploys it to your GPU server. For this to work, Claude Code needs to know your server environment.

Three GPU modes are supported — pick one and add it to your project's CLAUDE.md:

Option A: Remote SSH Server (gpu: remote)

## Remote Server
- gpu: remote
- SSH: `ssh my-gpu-server` (key-based auth, no password)
- GPU: 4x A100
- Conda env: `research` (Python 3.10 + PyTorch)
- Activate: `eval "$(/opt/conda/bin/conda shell.bash hook)" && conda activate research`
- Code directory: `/home/user/experiments/`
- Use `screen` for background jobs: `screen -dmS exp0 bash -c '...'`

Claude Code reads this and knows how to SSH in, activate the environment, and launch experiments. GPT-5.4 (the reviewer) only decides what experiments to run — Claude Code figures out how based on your CLAUDE.md.

Option B: Local GPU (gpu: local)

If you are already on the GPU server, you can add the following to your CLAUDE.md:

## GPU Environment
- gpu: local
- This machine has direct GPU access (no SSH needed)
- GPU: 4x A100 80GB
- Experiment environment: `YOUR_CONDA_ENV` (Python 3.x + PyTorch)
- Activate before any Python command: `The command to activate your experiment environment` (uv, conda, etc.)
- Code directory: `/home/YOUR_USERNAME/YOUR_CODE_DIRECTORY/`

Option C: Vast.ai On-Demand GPU (gpu: vast)

No GPU? Rent one from Vast.ai on demand. ARIS analyzes your training task (model size, dataset, time), finds the cheapest GPU that fits, ranks options by total cost (not just $/hr), then rents → runs → collects → destroys automatically.

Drop this in CLAUDE.md:

## Vast.ai
- gpu: vast                  # rent on-demand GPU from vast.ai
- auto_destroy: true         # auto-destroy after experiment completes (default)
- max_budget: 5.00           # optional: warn if estimated cost exceeds this

📖 Full setup guide → docs/integrations/VAST_GPU_GUIDE.md covers:

• Account + vastai CLI install + API key + SSH key prerequisites (5 steps)
• How ARIS picks GPUs and shows a live cost-ranked table
• Manual rental via /vast-gpu (list / rent / destroy)
• Typical cost expectations (~$0.30-2 for ablations on RTX 4090, ~$2-10 for A100/H100 baselines)
• When gpu: vast is preferable to gpu: remote / gpu: local

No server at all? The review and rewriting skills still work without GPU access. Only experiment-related fixes will be skipped (flagged for manual follow-up).

Connect your Zotero library to /research-lit — search collections, read your annotations/highlights, export BibTeX, all before web search. Recommended MCP: zotero-mcp (1.8k⭐, semantic search + PDF annotations + BibTeX export).

📖 Full setup guide → docs/integrations/ZOTERO.md covers:

• zotero-mcp install (Local API for desktop, or Web API)
• API key + user ID setup
• What it enables in /research-lit (semantic search, collections, PDF annotations, BibTeX export)
• New default source order: Zotero → Obsidian → local PDFs → web
• Combined Zotero + Obsidian workflow

Not using Zotero? /research-lit auto-skips and uses local PDFs + web search instead.

Connect your Obsidian vault to /research-lit — search your notes, tagged references, and processed insights (often more valuable than raw papers). Recommended MCP: mcpvault (760⭐, no Obsidian app needed). Pairs with Zotero. arXiv access is built-in — no setup needed, /research-lit queries the arXiv API automatically.

📖 Full setup guide → docs/integrations/OBSIDIAN.md covers:

• mcpvault install (point to vault path, BM25 search, 14 tools)
• Optional obsidian-skills (13.6k⭐, by Obsidian CEO) for wikilinks/callouts
• What it enables in /research-lit (vault search, tag filter, processed summaries, wikilink traversal)
• Combined Zotero + Obsidian workflow
• arXiv defaults + how to enable PDF download (— arxiv download: true)
• Standalone /arxiv "topic" and /arxiv 2301.07041 — download

Not using Obsidian? /research-lit auto-skips and works as before. arXiv stays on regardless.

Get mobile notifications when experiments finish, reviews score, or checkpoints need your input — without sitting in front of the terminal.

Three modes — choose per-project:

📖 Full setup guide → docs/integrations/FEISHU.md covers:

• Push-only setup (5 min) — create a group bot, copy the webhook URL, drop ~/.claude/feishu.json, test with a curl
• Interactive setup (15 min) — Feishu Open Platform app, the 5 required permissions (incl. the easy-to-miss im:message.p2p_msg:readonly), feishu-claude-code bridge install, screen deployment
• Card color/content table (Review ≥ 6 → green, < 6 → orange, error → red, etc.)
• Which skills send notifications, push vs. interactive payload per skill
• Common bot-doesn't-reply troubleshooting table
• Alternative IM platforms (cc-connect, clawdbot-feishu, lark-openapi-mcp)

Not using Feishu? Without ~/.claude/feishu.json, all skills behave exactly as before. Zero overhead, zero side effects.

🎛️ Customization

Skills are plain Markdown files. Fork and customize:

💡 Parameter pass-through: Parameters flow down the call chain automatically. For example, /research-pipeline "topic" — sources: zotero, arxiv download: true passes sources and arxiv download through idea-discovery all the way down to research-lit. This also works for optional sources such as deepxiv and exa: /research-pipeline "topic" — sources: all, deepxiv, exa. You can set any downstream parameter at any level — just add — key: value to your command.

research-pipeline  ──→  idea-discovery      ──→  research-lit
                   ──→  experiment-bridge    ──→  run-experiment
                   ──→  auto-review-loop
                                            ──→  idea-creator
                                            ──→  novelty-check
                                            ──→  research-review

Full Research Pipeline (research-pipeline)

Tune end-to-end behavior: GPU target, arXiv download, code review, human checkpoints, base repo, W&B logging, compact summaries, reference paper, illustration backend, and auto-proceed.

Override inline: /research-pipeline "topic" — auto proceed: false, illustration: mermaid

Auto Review Loop (auto-review-loop)

Tune stopping criteria: how many review→fix iterations, score threshold to declare submission-ready, and GPU-hour budget above which long experiments get flagged for manual follow-up.

Idea Discovery (idea-discovery / idea-creator)

Tune the pilot phase: max hours per pilot, hard timeout, max ideas piloted in parallel, total GPU budget, plus auto-proceed and arXiv download toggles.

Override inline: /idea-discovery "topic" — pilot budget: 4h per idea, sources: zotero, arxiv download: true

Experiment Bridge (experiment-bridge)

Tune deployment safety: GPT-5.4 code review, auto-deploy after review, sanity-test smallest experiment first, parallel run cap, W&B logging, and base-repo URL.

Override inline: /experiment-bridge — base repo: https://github.com/org/project

Literature Search (research-lit)

Tune sourcing: local PDF directories, local-scan cap, which sources to search (Zotero / Obsidian / web / Semantic Scholar / DeepXiv / Exa), and arXiv PDF download settings.

Override inline: /research-lit "topic" — sources: zotero, web, /research-lit "topic" — sources: all, deepxiv, /research-lit "topic" — sources: all, exa, /research-lit "topic" — arxiv download: true, max download: 10

Paper Writing (paper-write)

Tune paper format: real BibTeX from DBLP, target venue (ICLR/NeurIPS/ICML/CVPR/ACL/AAAI/IEEE…), anonymous author block, page limit, and illustration backend.

Override inline: /paper-write — target venue: NeurIPS, illustration: mermaid

General (all skills using Codex MCP)

Tune the reviewer model used by every Codex MCP call (default gpt-5.5), or fork the SKILL.md to customize prompt templates and the per-skill tool allowlist.

• Prompt templates — tailor the review persona and evaluation criteria
• allowed-tools — restrict or expand what each skill can do

🔀 Alternative Model Combinations

Don't have Claude / OpenAI API access? You can swap in other models — same cross-model architecture, different providers.

⭐ We strongly recommend Claude + GPT-5.4 (default setup). It's the most tested and reliable combination. Alternative setups work but may require prompt tuning.

Beyond the default Claude × GPT-5.4 route, ARIS ships 9 alternative routes (Alt A-I) covering Z.ai's GLM, Alibaba's Kimi/Qwen/GLM/MiniMax bundle, ModelScope's free DeepSeek-V3.1, Codex-as-executor with Claude or Gemini reviewers, and Google Antigravity as the executor.

How to choose:

• Default — you have Claude + OpenAI access and want the most tested route.
• Alt A — only swap Claude for GLM, keep GPT-5.4 as reviewer via Codex MCP.
• Alt B or Alt E — no Claude API, no OpenAI API (Alt E is free via ModelScope).
• Alt C or Alt D — OpenAI-compatible mix-and-match (Alt D = one Alibaba key for both sides).
• Alt G or Alt I — Codex stays as executor, only the reviewer changes (Claude or Gemini).
• Alt H — Antigravity as the executor (Claude Opus 4.6 or Gemini 3.1 Pro), GPT-5.4 or any llm-chat reviewer.

* Alt G normally relies on local Codex CLI and Claude Code CLI logins. Direct API keys are optional, not required.

Alt C supports tested providers: GLM (Z.ai), Kimi (Moonshot), LongCat (Meituan) as executors; DeepSeek, MiniMax as reviewers. Any OpenAI-compatible API should also work via the generic llm-chat MCP server.

Alt D uses Alibaba Coding Plan — one API key for both executor and reviewer, 4 models included (Kimi, Qwen, GLM, MiniMax).

Alt E uses ModelScope — free (2000 calls/day), one key, no automation restrictions.

Alt G keeps Codex as executor but swaps the reviewer to Claude Code CLI via the local claude-review MCP bridge, with async polling for long paper/review prompts.

Alt H uses Google Antigravity as the executor with native SKILL.md support — choose Claude Opus 4.6 (Thinking) or Gemini 3.1 Pro (high) as the execution model.

Alt I keeps Codex as executor, adds only a thin skills-codex-gemini-review overlay, and routes the reviewer-aware predefined skills through the local gemini-review MCP bridge with direct Gemini API by default. It is the closest Gemini analogue to the existing Codex+Claude review path, while minimizing skill changes and now also covers poster PNG review via the same bridge. Free-tier availability, rate limits, and data-use terms remain subject to Google's current policy.

Alt A: GLM + GPT

Only replace the executor (Claude → GLM via Z.ai), keep GPT-5.4 as reviewer via Codex MCP. Codex CLI reuses your existing OPENAI_API_KEY (from ~/.codex/config.toml or environment); no extra reviewer-side config.

npm install -g @anthropic-ai/claude-code
npm install -g @openai/codex
codex setup   # set model to gpt-5.5

Configure ~/.claude/settings.json:

{
    "env": {
        "ANTHROPIC_AUTH_TOKEN": "your_zai_api_key",
        "ANTHROPIC_BASE_URL": "https://api.z.ai/api/anthropic",
        "API_TIMEOUT_MS": "3000000",
        "ANTHROPIC_DEFAULT_HAIKU_MODEL": "glm-4.5-air",
        "ANTHROPIC_DEFAULT_SONNET_MODEL": "glm-4.7",
        "ANTHROPIC_DEFAULT_OPUS_MODEL": "glm-5"
    },
    "mcpServers": {
        "codex": {
            "command": "/opt/homebrew/bin/codex",
            "args": ["mcp-server"]
        }
    }
}

Alt B: GLM + MiniMax

No Claude or OpenAI API needed. Uses a custom MiniMax MCP server instead of Codex (because MiniMax doesn't support OpenAI's Responses API). Full guide: docs/MINIMAX_MCP_GUIDE.md.

Alt C: Any Executor + Any Reviewer

Mix and match freely using the generic llm-chat MCP server. Supports any OpenAI-compatible API as reviewer. Full guide: docs/LLM_API_MIX_MATCH_GUIDE.md.

Example combinations: GLM + DeepSeek, Kimi + MiniMax, Claude + DeepSeek, LongCat + GLM, etc.

After Setup: Install Skills & Verify

Use the project-local symlink install from § Install Skills above — that's the recommended path for all routes. The global-copy fallback below also works if you prefer everything under ~/.claude/skills/.

git clone https://github.com/wanshuiyin/Auto-claude-code-research-in-sleep.git
cd Auto-claude-code-research-in-sleep
cp -r skills/* ~/.claude/skills/
claude

⚠️ For non-Claude executors (GLM, Kimi, etc.): Let the model read through the project once to ensure skills are correctly parsed. This is especially important if you've rewritten skills to use a different reviewer MCP (e.g., mcp__llm-chat__chat instead of mcp__codex__codex) — the new executor needs to understand the changed tool call patterns:

Read through this project and verify all skills are working:
/idea-creator, /research-review, /auto-review-loop, /novelty-check,
/idea-discovery, /research-pipeline, /research-lit, /run-experiment,
/analyze-results, /monitor-experiment, /pixel-art

⚠️ Note: Alternative models may behave differently from Claude and GPT-5.4. You may need to tune prompt templates for best results. The core cross-model architecture remains the same.

💬 Community

Domain-specific skills welcome! The core skills cover general research workflows, but every field has its own tools and patterns. We welcome PRs that add new skills for your domain — EDA, bioinformatics, robotics, HPC, or anything else. Just add a skills/your-skill/SKILL.md and open a PR. See dse-loop for an example.

Join the WeChat group for discussion on Claude Code + AI-driven research workflows:

📖 Citation

If you use ARIS in your research, please cite:

@article{yang2026aris,
  title={ARIS: Autonomous Research via Adversarial Multi-Agent Collaboration},
  author={Yang, Ruofeng and Li, Yongcan and Li, Shuai},
  journal={arXiv preprint arXiv:2605.03042},
  year={2026}
}

⭐ Star History

🙏 Acknowledgements

Inspired by — 🧪 AI Scientist (Sakana) · 📖 AutoResearch (Karpathy) · 🔭 FARS (Analemma) · 🎨 PaperBanana (PKU).

Core infra — Claude Code (execution backbone) · Codex CLI (cross-model review via MCP).

Integrations — Zotero (guide): zotero-mcp, Zotero. Obsidian (guide): mcpvault, obsidian-skills (by Obsidian CEO Steph Ango). Feishu/Lark (guide): feishu-claude-code, clawdbot-feishu, cc-connect, lark-openapi-mcp.

Paper-writing inspiration — claude-scholar · Research-Paper-Writing-Skills · baoyu-skills. Community — awesome-agent-skills (featured).

Platform adaptation — 🤖 @Falling-Flower (Codex CLI adaptation via spawn_agent) · 🔧 @No-518 (Codex skill maintenance) · 🖱️ @YecanLee (Cursor guide + local GPU docs) · 🏆 @DefanXue & @Monglitay (first ARIS community paper, CS conference 8/10).

Architecture & vision — 💡 @JingxuanKang: beyond code (training-check, result-to-claim, ablation-planner, watchdog, templates, session recovery), deeply shaped ARIS through discussions on compact mode, workflow state management, and the vision of autonomous research — many of today's core features (structured project files, context-aware session recovery) grew out of these conversations.

License

MIT