自睡眠自研究系统

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

<p align="center"> <a href="https://huggingface.co/papers/2605.03042"> <img src="docs/hf_daily_paper_1.svg" alt="Hugging Face Daily Paper · #1 Paper of the Day" width="360"> </a> </p>

· · · · · · -orange?style=flat) · · 💬 Join Community ·

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

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

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

🛡️ ARIS audits its own output → now Anti-Autoresearch audits everyone's. It catalogs 39 autoresearch hack-patterns across 7 families and checks a submission for them end-to-end, producing a deterministic, reviewer-ready integrity report. Self-consistency + fabrication forensics, not an AI-text detector.

<p align="center"><em>The field has put up with unreliable autoresearch long enough —<br>Anti-Autoresearch is the read that finally catches it.</em></p>

🎬 ARIS goes multimodal → ARIS-Movie-Director — hand over a fuzzy story, wake up to a cross-model-audited movie (reference run = 19 scenes). Long-horizon visual stories drift two ways (🧠 long-range forgetting · 🗣️ each frame signed off by the model that drew it); ARIS answers with the same DNA — a research-wiki for memory + multi-agent debate so no frame signs off on itself.

<p align="center"> <a href="https://wanshuiyin.github.io/ARIS-Movie-Director/comic/"> <img src="https://raw.githubusercontent.com/wanshuiyin/ARIS-Movie-Director/main/docs/comic_cover_v4.webp" alt="ARIS-Movie-Director — watch the cross-model-audited image movie (19 scenes) in your browser" width="100%"> </a> </p>

<p align="center"> <a href="https://github.com/wanshuiyin/ARIS-Movie-Director"> <img src="docs/aris-movie-director-method.png" alt="ARIS-Movie-Director method — the audited spiral: authored source of truth (asset library · outline · storyboard · comic.json) → per-panel image_gen + cross-model panel_gate (blind token-diff, single-vote veto) → research-wiki audit trace → assembly + release" width="100%"> </a> </p>

🧭 Not just movies — the same audited spiral also generates clean method / flow diagrams: this very figure was baked by ARIS-Movie-Director's image_gen + cross-model panel_gate loop. 👉 Skills + an end-to-end CLI in ARIS-Movie-Director: /movie-pipeline (agent workflow + standalone deterministic CLI core) and /method-figure, the skill that made this figure.

<details> <summary>🎞️ <i>A few frames from the reference movie — the story's own integrity beat: a run that <b>reported <code>+6.2</code></b> but <b>really moved <code>+1.4</code></b>.</i> &nbsp;<b><a href="https://wanshuiyin.github.io/ARIS-Movie-Director/comic/">▶ watch all 19 scenes →</a></b></summary>

</details>

🎯 准备 2026 AI 秋招？ → 🌐 ARIS-in-AI-Offer · GitHub repo · 中文 README —— 23 篇双语 ML / LLM / 多模态 / 生成式 / Agent 面试 cheat sheet，每篇 = 公式推导 + 从零 PyTorch + 25 高频面试题（L1 / L2 / L3），全部由 ARIS 的 /render-html 自动生成。希望大家秋招轻松一点 🌱

<details> <summary><b>🖼️ Preview</b> — the three-pillar cheat-sheet strip (① Foundations · ② Interview Q&amp;A · ③ From-Scratch Code)</summary>

<p align="center"> <a href="https://github.com/wanshuiyin/ARIS-in-AI-Offer"> <img src="https://raw.githubusercontent.com/wanshuiyin/ARIS-in-AI-Offer/main/assets/preview_strip.jpg" alt="ARIS-in-AI-Offer preview — ① Foundations + ② Interview Q&A + ③ From-Scratch Code, three columns from a representative cheat sheet" width="100%"> </a> </p>

</details>

📖 Preview from the Diffusion Foundations cheat sheet — every tutorial in ARIS-in-AI-Offer follows the same three-pillar structure (foundations / interview Q&A / runnable code). 🌐 Same workflow, different deliverable — ARIS-Homepage v1 live demo (CV → fact-checked single-file academic homepage via /homepage-generator). 📝 Three long-form blogs, cross-model collaborative writing via /render-html — Continuous DLM — a representation-perspective survey (2026 H1) · Cosmos 3 — understanding + generation in one Transformer (MoT) · Diffusion × representation × manifold learning.

🛰 社区好物 · Claude Fleet（by @tianyilt）—— 本地只读看板，一眼盯住并行的一堆 Claude Code / Codex 窗口(谁在跑 / 等授权 / 跑完了)，一键跳转 + 全文搜 transcript。多 agent 工作流神器，好用点个 ⭐

🪟 更轻的自家选择 · ARIS-Monitor —— ARIS 自带的 macOS 置顶悬浮小窗(纯 Python · 无浏览器)：只亮"哪个会话在等你授权" 🔴，点一行跳到那个终端。

<details> <summary><b>🖼️ Preview</b> — Claude Fleet dashboard (full web) &amp; ARIS-Monitor widget (minimal, built-in)</summary>

Claude Fleet · 全功能网页看板	ARIS-Monitor · 极简悬浮小窗(自带)

</details>

<details> <summary><b>Run either in seconds</b> — ARIS-Monitor (5s) / Claude Fleet (30s)</summary>

ARIS-Monitor — built-in, no clone / no pip / no browser:

```bash cd aris-monitor && ./run.sh

2. 📢 What's New

- 2026-06-20 — 📚 Research wiki: all four node layers now have deterministic writers — fixes "re-generated ideas not recorded" (#305, #306, #307, #308). A user hit a real bug — ideas recorded on the first /idea-creator run vanished on re-generation — because wiki pages were written freehand, a prose step the model skips on a re-prompt. Each layer now has a dedicated research_wiki.py writer joining ingest_paper: add_claim (claims born at /proof-checker), upsert_idea (/idea-creator), add_experiment (/result-to-claim) — each guarded by a drift-check so it can't silently regress to dead code. A claim's status is now a strict proof axis (verified/refuted/unproven/…) while experiment support is carried by supports/invalidates edges (closing a latent contradiction the shared validator rejected), and the Codex-CLI skill mirror is synced to match. Zero behavior change when no research-wiki/ is present. - 2026-06-19 — 🛰 Overnight-loop resilience: silent-death watchdog + stall→structural-pivot (#300, #301, #302; operational patterns absorbed from Deli Chen's AutoResearch framework). Two failure modes an unattended /loop / CronCreate heartbeat couldn't catch. (1) Silent death — the heartbeat is parasitic on a living session, so context compaction or a closed session kills it and nothing notices. A new watchdog loop task type (watchdog.py) judges liveness by the state file's mtime against the loop's own stale_after_seconds, surfacing STALE / MISSING / COMPLETED to alerts.log — detect-only, it never restarts a verdict-bearing loop. (2) Cognitive spin — a stalled loop retries near-variants forever. A new iteration_log.py counts NEW findings per tick: stale_count ≥ 2 forces a structural pivot (change the frame + pick an untried direction), ≥ 4 escalates to a human. Both are Type-A signals — "keep going / change direction," never "good enough"; quality still terminates in the cross-model jury. - 2026-06-07 — 🖼️ /paper-poster-html — new DEFAULT poster pipeline (skill #79); LaTeX /paper-poster retired to a redirect stub. Builds the poster as a single HTML/CSS file on the venue's exact print canvas and iterates by measuring, not eyeballing: hard gates (column-balance spread < 5 px, two-hue design-token discipline, real-paper-figure provenance manifest, figure-area bands) must PASS before any reviewer sees the poster; a closed fix vocabulary (token / component / rebalance / asset / canvas) structurally kills the cosmetic patch-loop; a fresh cross-model review acquits content fidelity (claim→evidence audit + final print-readiness pass). Ships 3 templates + a catalogued component library (incl. density components: equation anatomy, flow-strip, duo figures, derived-Δ tables, claim pills) and 6 venue token packs. Core gate machinery adapted from posterly (MIT, by @Chenruishuo) — ARIS adds the style/asset gates, the density system, and the cross-model loop. ⚠️ /paper-poster now redirects to /paper-poster-html; the legacy LaTeX pipeline remains only in git history. - 2026-05-31 — 🤝 Community spotlight — two tools worth a look. Claude Fleet (@tianyilt) — a local read-only dashboard to triage / Focus / full-text-search across many concurrent Claude Code + Codex windows. posterly (@Chenruishuo) — a Claude Code skill that builds academic conference posters as a single HTML/CSS file → print-ready PDF via headless Chromium (no LaTeX). Both indexed under Awesome Community. 🌟 if they help you. <details> <summary>Earlier updates (2026-03-12 — 2026-05-31, 71 entries)</summary>

• 2026-05-31 — 🛰 Fourth reviewer backend: Gemini via Antigravity CLI (#267 by @ZGJY95). — reviewer: agy routes review through the Antigravity CLI for users without Codex MCP / Oracle — fail-closed on the cross-model invariant (recovers + verifies the real Gemini-family model, refuses non-Gemini, binds the recovered transcript to the call via a user-event nonce). Wired into reviewer-routing.md.

• 2026-05-29 — ⚙️ ultracode-native convention layer — fan out for breadth on any runtime tier, keep the cross-model jury sacred. Three new shared-references docs decouple breadth from verdict: fan-out-pattern.md (skills generate candidates across same-family Claude subagents — Tier-1 Workflow / Tier-2 Agent / Tier-3 sequential — all ending in the identical cross-model jury), acceptance-gate.md ("a loop can DRIVE, it cannot ACQUIT" — self-judge execution-completeness, never quality/correctness), and external-cadence.md (/loop & CronCreate are fire-control, never a jury). Wired into /idea-creator, /research-lit, /proof-checker, /kill-argument (fan-out) plus 16 skills (cadence fence/affordance). Also stripped 48 vestigial Agent grants (least-privilege + a drift-check guard), fixed /idea-creator's same-family idea pre-filter, and reconciled an /auto-review-loop OR→AND stop-condition inconsistency. Non-ultracode users benefit immediately — fan-out degrades to sequential with the same final jury.

• 2026-05-28 — 📝 First blog shipped: A Survey on Continuous DLM (2026 H1, 6 papers) — long-form bilingual technical survey by Ruofeng Yang (SJTU), written end-to-end through the ARIS-in-AI-Offer workflow (Claude Opus 4.7 + Codex GPT-5.5 xhigh + Gemini auto-gemini-3 cross-model discussion). Compares ELF, ByteDance Cola-DLM, and Flow-Matching family across discrete-DLM problems, the "known-unknown" continuous space idea, training pipeline, architecture / params / shapes, inference grids + Tab 6/7 numerical results, denoising trajectories, and a Field Landscape against Cola-DLM. A 1.7 MB self-contained HTML (no build) — demonstrates the kind of long-form analysis the /render-html toolchain can produce.

• 2026-05-26 — 🌐 HTML auto-emission activated at 8 workflow checkpoints. /idea-discovery, /auto-review-loop, /research-pipeline, /kill-argument, /proof-checker, /paper-claim-audit, /citation-audit, /rebuttal now auto-render their primary MD artifact to a single-file HTML view via /render-html. Cost-tiered: interim views use --no-review, audit-class / reviewer-facing deliverables keep the full Codex render-fidelity gate. Default on (RENDER_HTML = true); per-skill opt-out. Failures non-blocking — source MD stays canonical.
• 2026-05-26 — 🤝 Community PR wave — 5 merges this week. /wiki-enrich (#247 by @hungchun0201) fills paper TODOs ingest_paper leaves as scaffolds — Karpathy LLM-wiki principle, fetch chain alphaxiv→deepxiv→arXiv. Mirror drift checker + CI (#241 by @VeraPyuyi) keeps main↔mirror in sync. /research-pipeline Stage 2/3 unified into /experiment-bridge delegation (#243 by @ZBigFish) — old inline was a strict subset of the bridge. Windows PowerShell installer parity with reparse-chain inside-repo guard + -FromOld legacy migration + Windows CI matrix (#242 by @VeraPyuyi). Plus manual-review MCP (#246 by @ZBigFish) — third reviewer backend — reviewer: manual for zero-cost cross-model review (paste prompt to any non-Claude model: DeepSeek / Kimi / ChatGPT / Gemini / local llama); cross-model invariant guarded by bilingual UI banner + per-session token auth + fail-closed when MCP unavailable.

• 2026-05-17 — 🐙 GitHub Copilot CLI adaptation — native SKILL.md + MCP support, no skill mirror needed. Installer (install_aris_copilot.sh) + smart-updater + 13-test suite. Community contribution by @EarendelH (#229, closes #214 / #227 / #203).
• 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 + ` discipline** ([#217](https://github.com/wanshuiyin/Auto-claude-code-research-in-sleep/issues/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 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_pape