AutoResearchClaw

Chat an Idea. Get a Paper. Autonomous, Collaborative & Self-Evolving.

Just chat with OpenClaw: "Research X" → done.

📄 Our paper is on arXiv — come read it! AutoResearchClaw: Self-Reinforcing Autonomous Research with Human-AI Collaboration

🇨🇳 中文 · 🇯🇵 日本語 · 🇰🇷 한국어 · 🇫🇷 Français · 🇩🇪 Deutsch · 🇪🇸 Español · 🇧🇷 Português · 🇷🇺 Русский · 🇸🇦 العربية

🏆 Paper Showcase · 🧑‍✈️ Co-Pilot Guide · 📖 Integration Guide · 💬 Discord Community

🏆 Generated Paper Showcase 8 papers across 8 domains — math, statistics, biology, computing, NLP, RL, vision, robustness — generated fully autonomously or with Human-in-the-Loop co-pilot guidance.

🧪 We're looking for testers! Try the pipeline with your own research idea — from any field — and tell us what you think. Your feedback directly shapes the next version. → Testing Guide | → 中文测试指南 | → 日本語テストガイド

🔥 News

• [05/19/2026] v0.5.0 — Multi-Domain Experiment Agents + ARC-Bench — Two headline updates. (1) Domain-specialist execution agents: the experiment stage (Stages 10–13) now routes beyond the default ML sandbox to specialist agents per field — high-energy physics (ColliderAgent: Lagrangian → FeynRules → MadGraph5 → Delphes via the Magnus cloud), biology (COBRApy genome-scale metabolic modelling), and statistics (simulation-study agent), with a generic Docker executor covering chemistry/materials. The pipeline auto-selects the right executor from the research domain. (2) ARC-Bench: a 55-topic open-ended autonomous-research benchmark spanning ML (25), HEP (10), quantum (10), biology (7), and statistics (3) — each topic ships a manifest (research question + conditions + metrics + datasets) and a rubric for graded scoring, all under experiments/arc_bench/, and also released on 🤗 Hugging Face. → Domain Integration Guide
• [04/01/2026] v0.4.0 — Human-in-the-Loop Co-Pilot System — AutoResearchClaw is no longer purely autonomous. New HITL system adds 6 intervention modes (full-auto, gate-only, checkpoint, step-by-step, co-pilot, custom), per-stage policies, and deep human-AI collaboration. Includes: Idea Workshop for hypothesis co-creation, Baseline Navigator for experiment design review, Paper Co-Writer for collaborative drafting, SmartPause (confidence-driven dynamic intervention), ALHF intervention learning, anti-hallucination claim verification, cost budget guardrails, pipeline branching for parallel hypothesis exploration, and CLI commands (attach/status/approve/reject/guide). → Full HITL Guide
• [03/30/2026] Flexible Skill Loading — AutoResearchClaw now supports loading open-source and custom skills from any discipline to further enhance your research experience. 20 pre-loaded skills are included as ready-to-use references, covering scientific writing, experiment design, chemistry, biology, and more — including an A-Evolve agentic evolution skill contributed by the community. Load your own via researchclaw skills install or drop a SKILL.md into .claude/skills/. See Skills Library.
• [03/22/2026] v0.3.2 — Cross-Platform Support + Major Stability — AutoResearchClaw now runs on any ACP-compatible agent backend (Claude Code, Codex CLI, Copilot CLI, Gemini CLI, Kimi CLI) and supports messaging platforms (Discord, Telegram, Lark, WeChat) via OpenClaw bridge. New CLI-agent code generation backend delegates Stages 10 & 13 to external CLI agents with budget control and timeout management. Also includes anti-fabrication system (VerifiedRegistry + experiment diagnosis & repair loop), 100+ bug fixes, modular executor refactoring, --resume auto-detection, LLM retry hardening, and community-reported fixes.

⚡ One Command. One Paper.

# Fully autonomous — no human intervention
pip install -e . && researchclaw setup && researchclaw init && researchclaw run --topic "Your research idea here" --auto-approve

# Co-Pilot mode — collaborate with AI at key decision points
researchclaw run --topic "Your research idea here" --mode co-pilot

🤔 What Is This?

You think it. AutoResearchClaw writes it. You guide the key decisions.

Drop a research topic — get back a full academic paper with real literature from OpenAlex, Semantic Scholar & arXiv, hardware-aware sandbox experiments (GPU/MPS/CPU auto-detected), statistical analysis, multi-agent peer review, and conference-ready LaTeX targeting NeurIPS/ICML/ICLR. Run it fully autonomous, or use Co-Pilot mode to guide the AI at critical decision points — choose research directions, review experiment designs, and co-write the paper. No hallucinated references.

The pipeline runs end-to-end — fully autonomous or with human-in-the-loop collaboration. When experiments fail, it self-heals. When hypotheses don't hold, it pivots. When citations are fake, it kills them. When you want to steer, it pauses and listens.

🌍 Run it anywhere. AutoResearchClaw isn't locked to a single platform. Use it standalone via CLI, plug it into OpenClaw, or wire it up through any ACP-compatible agent — 🤖 Claude Code, 💻 Codex CLI, 🐙 Copilot CLI, ♊ Gemini CLI, 🌙 Kimi CLI, you name it. And because OpenClaw bridges to messaging platforms, you can kick off a full research run from 💬 Discord, ✈️ Telegram, 🐦 Lark (飞书), 💚 WeChat, or wherever your team already hangs out. One topic in, one paper out — no matter where you type it.

🚀 Quick Start

# 1. Clone & install
git clone https://github.com/aiming-lab/AutoResearchClaw.git
cd AutoResearchClaw
python3 -m venv .venv && source .venv/bin/activate
pip install -e .

# 2. Setup (interactive — installs OpenCode beast mode, checks Docker/LaTeX)
researchclaw setup

# 3. Configure
researchclaw init          # Interactive: choose LLM provider, creates config.arc.yaml
# Or manually: cp config.researchclaw.example.yaml config.arc.yaml

# 4. Run
export OPENAI_API_KEY="sk-..."
researchclaw run --config config.arc.yaml --topic "Your research idea" --auto-approve

Output → artifacts/rc-YYYYMMDD-HHMMSS-<hash>/deliverables/ — compile-ready LaTeX, BibTeX, experiment code, charts.

project:
  name: "my-research"

research:
  topic: "Your research topic here"

llm:
  base_url: "https://api.openai.com/v1"
  api_key_env: "OPENAI_API_KEY"
  primary_model: "gpt-4o"
  fallback_models: ["gpt-4o-mini"]

experiment:
  mode: "sandbox"
  sandbox:
    python_path: ".venv/bin/python"

🧠 What Makes It Different

🦞 OpenClaw Integration

AutoResearchClaw is an OpenClaw-compatible service. Install it in OpenClaw and launch autonomous research with a single message — or use it standalone via CLI, Claude Code, or any AI coding assistant.

🚀 Use with OpenClaw (Recommended)

If you already use OpenClaw as your AI assistant:

1️⃣  Share the GitHub repo URL with OpenClaw
2️⃣  OpenClaw auto-reads RESEARCHCLAW_AGENTS.md → understands the pipeline
3️⃣  Say: "Research [your topic]"
4️⃣  Done — OpenClaw clones, installs, configures, runs, and returns results

That's it. OpenClaw handles git clone, pip install, config setup, and pipeline execution automatically. You just chat.

🔌 OpenClaw Bridge (Advanced)

For deeper integration, AutoResearchClaw includes a bridge adapter system with 6 optional capabilities:

# config.arc.yaml
openclaw_bridge:
  use_cron: true              # ⏰ Scheduled research runs
  use_message: true           # 💬 Progress notifications (Discord/Slack/Telegram)
  use_memory: true            # 🧠 Cross-session knowledge persistence
  use_sessions_spawn: true    # 🔀 Spawn parallel sub-sessions for concurrent stages
  use_web_fetch: true         # 🌐 Live web search during literature review
  use_browser: false          # 🖥️ Browser-based paper collection

Each flag activates a typed adapter protocol. When OpenClaw provides these capabilities, the adapters consume them without code changes. See docs/integration-guide.md for full details.

ACP (Agent Client Protocol)

AutoResearchClaw can use any ACP-compatible coding agent as its LLM backend — no API keys required. The agent communicates via acpx, maintaining a single persistent session across all 23 pipeline stages.

# config.yaml — ACP example
llm:
  provider: "acp"
  acp:
    agent: "claude"   # Any ACP-compatible agent CLI command
    cwd: "."          # Working directory for the agent
  # No base_url or api_key needed — the agent handles its own auth.

# Just run — the agent uses its own credentials
researchclaw run --config config.yaml --topic "Your research idea" --auto-approve

🛠️ Other Ways to Run

🔬 Pipeline: 23 Stages, 8 Phases

Phase A: Research Scoping          Phase E: Experiment Execution
  1. TOPIC_INIT                      12. EXPERIMENT_RUN
  2. PROBLEM_DECOMPOSE               13. ITERATIVE_REFINE  ← self-healing

Phase B: Literature Discovery      Phase F: Analysis & Decision
  3. SEARCH_STRATEGY                 14. RESULT_ANALYSIS    ← multi-agent
  4. LITERATURE_COLLECT  ← real API  15. RESEARCH_DECISION  ← PIVOT/REFINE
  5. LITERATURE_SCREEN   [gate]
  6. KNOWLEDGE_EXTRACT               Phase G: Paper Writing
                                     16. PAPER_OUTLINE
Phase C: Knowledge Synthesis         17. PAPER_DRAFT
  7. SYNTHESIS                       18. PEER_REVIEW        ← evidence check
  8. HYPOTHESIS_GEN    ← debate      19. PAPER_REVISION

Phase D: Experiment Design         Phase H: Finalization
  9. EXPERIMENT_DESIGN   [gate]      20. QUALITY_GATE      [gate]
 10. CODE_GENERATION                 21. KNOWLEDGE_ARCHIVE
 11. RESOURCE_PLANNING               22. EXPORT_PUBLISH     ← LaTeX
                                     23. CITATION_VERIFY    ← relevance check

Gate stages (5, 9, 20) pause for human approval or auto-approve with --auto-approve. On rejection, the pipeline rolls back.

Co-Pilot mode (--mode co-pilot): Deep human-AI collaboration at Stages 7-8 (Idea Workshop), Stage 9 (Baseline Navigator), and Stages 16-17 (Paper Co-Writer). Other stages auto-execute with SmartPause monitoring.

Decision loops: Stage 15 can trigger REFINE (→ Stage 13) or PIVOT (→ Stage 8), with automatic artifact versioning.

✨ Key Features

🧑‍✈️ Human-in-the-Loop Co-Pilot

AutoResearchClaw v0.4.0 introduces a complete Human-in-the-Loop (HITL) system that transforms the pipeline from purely autonomous to a human-AI collaborative research engine. Choose your level of involvement:

Intervention Modes

Co-Pilot Workflow(updated Apr 13, added experiment to prove the best)

You: researchclaw run --topic "Quantum noise as neural network regularization" --mode co-pilot

Pipeline runs Stages 1-7 automatically...

  ┌─────────────────────────────────────────────────────────────┐
  │  HITL | Stage 08: HYPOTHESIS_GEN                            │
  │  Post-stage review                                          │
  │                                                             │
  │  Hypotheses mentioned: 3                                    │
  │  Novelty score: 0.72 (moderate)                             │
  │                                                             │
  │  [a] Approve  [r] Reject  [e] Edit  [c] Collaborate         │
  │  [i] Inject guidance  [v] View output  [q] Abort            │
  └─────────────────────────────────────────────────────────────┘

You: c  (start collaborative chat)
You: Hypothesis 3 is interesting but needs Dropout/Label Smoothing as baselines
AI:  Updated — added Dropout, Label Smoothing, MixUp, CutMix as baselines...
You: approve

Pipeline continues with your refined hypothesis...

CLI Commands

# Start with HITL mode
researchclaw run --topic "..." --mode co-pilot

# Attach to a paused pipeline (from another terminal)
researchclaw attach artifacts/rc-2026-xxx

# Check pipeline and HITL status
researchclaw status artifacts/rc-2026-xxx

# Approve/reject from another terminal or script
researchclaw approve artifacts/rc-2026-xxx --message "LGTM"
researchclaw reject artifacts/rc-2026-xxx --reason "Missing key baseline"

# Inject guidance for a stage (even before it runs)
researchclaw guide artifacts/rc-2026-xxx --stage 9 --message "Use ResNet-50 as primary baseline"

Key Capabilities

Configuration

# config.arc.yaml
hitl:
  enabled: true
  mode: co-pilot                     # full-auto | gate-only | checkpoint | co-pilot | custom
  cost_budget_usd: 50.0              # Pause when cost exceeds budget (0 = no limit)

  notifications:
    on_pause: true
    on_quality_drop: true
    channels: ["terminal"]            # terminal | slack | webhook

  timeouts:
    default_human_timeout_sec: 86400  # 24h default wait
    auto_proceed_on_timeout: false

  collaboration:
    max_chat_turns: 50
    save_chat_history: true

  # Per-stage custom policies (optional, for 'custom' mode)
  stage_policies:
    8: { require_approval: true, enable_collaboration: true }
    9: { require_approval: true, allow_edit_output: true }

Backward Compatibility

• Default: OFF. Without hitl.enabled: true or --mode, the pipeline behaves exactly as before.
• --auto-approve still works. It overrides HITL mode.
• All 2,699 existing tests pass with HITL code present.

🧠 MetaClaw Integration

AutoResearchClaw + MetaClaw = A pipeline that learns from every run.

MetaClaw adds cross-run knowledge transfer to AutoResearchClaw. When enabled, the pipeline automatically captures lessons from failures and warnings, converts them into reusable skills, and injects those skills into all 23 pipeline stages on subsequent runs — so the same mistakes are never repeated.

How It Works

Run N executes → failures/warnings captured as Lessons
                      ↓
          MetaClaw Lesson → Skill conversion
                      ↓
          arc-* Skill files stored in ~/.metaclaw/skills/
                      ↓
Run N+1 → build_overlay() injects skills into every LLM prompt
                      ↓
          LLM avoids known pitfalls → higher quality, fewer retries

Quick Setup

# 1. Install MetaClaw (if not already)
pip install metaclaw

# 2. Enable in your config

# config.arc.yaml
metaclaw_bridge:
  enabled: true
  proxy_url: "http://localhost:30000"        # MetaClaw proxy (optional)
  skills_dir: "~/.metaclaw/skills"          # Where skills are stored
  fallback_url: "https://api.openai.com/v1" # Direct LLM fallback
  fallback_api_key: ""                      # API key for fallback URL
  lesson_to_skill:
    enabled: true
    min_severity: "warning"                 # Convert warnings + errors
    max_skills_per_run: 3

# 3. Run as usual — MetaClaw works transparently
researchclaw run --config config.arc.yaml --topic "Your idea" --auto-approve

After each run, check ~/.metaclaw/skills/arc-*/SKILL.md to see the skills your pipeline has learned.

Experiment Results

In controlled A/B experiments (same topic, same LLM, same configuration):

Composite robustness score is a weighted average of stage completion rate (40%), retry reduction (30%), and refine cycle efficiency (30%).

Backward Compatibility

• Default: OFF. If metaclaw_bridge is absent or enabled: false, the pipeline behaves exactly as before.
• No new dependencies. MetaClaw is optional — the core pipeline works without it.
• All 2,699 existing tests pass with the integration code present.

🧩 Skills Library

AutoResearchClaw now supports loading open-source and custom skills to further enhance your research experience. We also ship with 20 pre-loaded built-in skills (scientific writing, literature search, chemistry, biology, and more) as ready-to-use references, offering a high degree of flexibility out of the box. Disable any skill by adding enabled: false to its frontmatter.

Sample built-in skills:

See all 20 skills with researchclaw skills list.

Load Your Own Skills

# Option 1: Install a skill (persists across projects)
researchclaw skills install /path/to/my-skill/

# Option 2: Drop a SKILL.md into the project
mkdir -p .claude/skills/my-custom-skill
# Then create a SKILL.md with YAML frontmatter (name, description, trigger-keywords, applicable-stages)

# Option 3: Configure shared skill directories in config.arc.yaml
# skills:
#   custom_dirs:
#     - /path/to/team-shared-skills

Using Skills

Skills are loaded and injected into LLM prompts automatically — no manual activation needed. Use the CLI to inspect:

researchclaw skills list               # Show all loaded skills with sources
researchclaw skills validate ./my-skill # Check SKILL.md format

Browse community skills: K-Dense-AI/claude-scientific-skills (150+ scientific skills across multiple disciplines).

⚙️ Configuration Reference

# === Project ===
project:
  name: "my-research"              # Project identifier
  mode: "docs-first"               # docs-first | semi-auto | full-auto

# === Research ===
research:
  topic: "..."                     # Research topic (required)
  domains: ["ml", "nlp"]           # Research domains for literature search
  daily_paper_count: 8             # Target papers per search query
  quality_threshold: 4.0           # Minimum quality score for papers

# === Runtime ===
runtime:
  timezone: "America/New_York"     # For timestamps
  max_parallel_tasks: 3            # Concurrent experiment limit
  approval_timeout_hours: 12       # Gate stage timeout
  retry_limit: 2                   # Retry count on stage failure

# === LLM ===
llm:
  provider: "openai-compatible"    # openai | openrouter | deepseek | minimax | acp | openai-compatible
  base_url: "https://..."          # API endpoint (required for openai-compatible)
  api_key_env: "OPENAI_API_KEY"    # Env var for API key (required for openai-compatible)
  api_key: ""                      # Or hardcode key here
  primary_model: "gpt-4o"          # Primary model
  fallback_models: ["gpt-4o-mini"] # Fallback chain
  s2_api_key: ""                   # Semantic Scholar API key (optional, higher rate limits)
  acp:                             # Only used when provider: "acp"
    agent: "claude"                # ACP agent CLI command (claude, codex, gemini, etc.)
    cwd: "."                       # Working directory for the agent

# === Experiment ===
experiment:
  mode: "sandbox"                  # simulated | sandbox | docker | ssh_remote
  time_budget_sec: 300             # Max execution time per run (default: 300s)
  max_iterations: 10               # Max optimization iterations
  metric_key: "val_loss"           # Primary metric name
  metric_direction: "minimize"     # minimize | maximize
  sandbox:
    python_path: ".venv/bin/python"
    gpu_required: false
    allowed_imports: [math, random, json, csv, numpy, torch, sklearn]
    max_memory_mb: 4096
  docker:
    image: "researchclaw/experiment:latest"
    network_policy: "setup_only"   # none | setup_only | pip_only | full
    gpu_enabled: true
    memory_limit_mb: 8192
    auto_install_deps: true        # Auto-detect imports → requirements.txt
  ssh_remote:
    host: ""                       # GPU server hostname
    gpu_ids: []                    # Available GPU IDs
    remote_workdir: "/tmp/researchclaw_experiments"
  opencode:                          # OpenCode Beast Mode (auto-installed via `researchclaw setup`)
    enabled: true                    # Master switch (default: true)
    auto: true                       # Auto-trigger without confirmation (default: true)
    complexity_threshold: 0.2        # 0.0-1.0 — higher = only trigger on complex experiments
    model: ""                        # Override model (empty = use llm.primary_model)
    timeout_sec: 600                 # Max seconds for OpenCode generation
    max_retries: 1                   # Retry count on failure
    workspace_cleanup: true          # Remove temp workspace after collection
  code_agent:                        # CodeAgent v2 — multi-phase code generation
    enabled: true                    # Use CodeAgent instead of legacy single-prompt codegen
    architecture_planning: true      # Generate deep implementation blueprint before coding
    sequential_generation: true      # Generate files one-by-one following dependency DAG
    hard_validation: true            # AST-based validation gates (blocks identical ablations, hardcoded metrics)
    hard_validation_max_repairs: 2   # Max repair attempts when validation fails
    exec_fix_max_iterations: 3       # Execution-in-the-loop fix attempts
    exec_fix_timeout_sec: 60         # Timeout per exec-fix attempt
  benchmark_agent:                   # BenchmarkAgent — automated dataset & baseline selection
    enabled: true                    # Enable 4-agent benchmark pipeline (Surveyor→Selector→Acquirer→Validator)
    enable_hf_search: true           # Search HuggingFace Datasets
    enable_web_search: true          # Search Google Scholar for benchmarks
    tier_limit: 2                    # Dataset tier filtering (1=small/cached, 2=medium, 3=large)
    min_benchmarks: 1                # Minimum datasets required
    min_baselines: 2                 # Minimum baseline methods required
  figure_agent:                      # FigureAgent — academic figure generation
    enabled: true                    # Enable 5-agent figure pipeline (Planner→CodeGen→Renderer→Critic→Integrator)
    min_figures: 3                   # Minimum figures to generate
    max_figures: 8                   # Maximum figures
    max_iterations: 3                # Critic-driven refinement iterations
    dpi: 300                         # Output resolution
    strict_mode: false               # Fail pipeline if figure generation fails
  repair:                            # Anti-fabrication experiment repair
    enabled: true                    # Auto-diagnose and repair failed experiments
    max_cycles: 3                    # Repair retry loops
    min_completion_rate: 0.5         # >=50% conditions must complete to proceed
    min_conditions: 2                # At least 2 conditions for valid experiment
    use_opencode: true               # Route repairs through OpenCode Beast Mode

# === Web Search (Optional) ===
web_search:
  enabled: true                      # Enable web-augmented literature search
  tavily_api_key_env: "TAVILY_API_KEY"  # Tavily API key env var (optional)
  enable_scholar: true               # Google Scholar search
  enable_pdf_extraction: true        # Extract text from PDFs
  max_web_results: 10                # Max web results per query

# === Export ===
export:
  target_conference: "neurips_2025"  # neurips_2025 | iclr_2026 | icml_2026
  authors: "Anonymous"
  bib_file: "references"

# === Prompts ===
prompts:
  custom_file: ""                  # Path to custom prompts YAML (empty = defaults)

# === HITL Co-Pilot (NEW in v0.4.0) ===
hitl:
  enabled: false                     # Set to true to enable HITL
  mode: co-pilot                     # full-auto | gate-only | checkpoint | step-by-step | co-pilot | custom
  cost_budget_usd: 0.0              # Cost limit in USD (0 = no limit)
  notifications:
    on_pause: true                   # Notify when pipeline pauses
    on_quality_drop: true            # Notify on quality issues
    channels: ["terminal"]           # terminal | slack | webhook
  timeouts:
    default_human_timeout_sec: 86400 # Wait up to 24h for human input
    auto_proceed_on_timeout: false   # If true, auto-approve on timeout
  collaboration:
    max_chat_turns: 50               # Max turns per collaboration session
    save_chat_history: true          # Persist chat logs
  stage_policies: {}                 # Per-stage overrides (for 'custom' mode)

# === Security ===
security:
  hitl_required_stages: [5, 9, 20] # Stages requiring human approval
  allow_publish_without_approval: false
  redact_sensitive_logs: true

# === Knowledge Base ===
knowledge_base:
  backend: "markdown"              # markdown | obsidian
  root: "docs/kb"

# === Notifications ===
notifications:
  channel: "console"               # console | discord | slack
  target: ""

# === MetaClaw Bridge (Optional) ===
metaclaw_bridge:
  enabled: false                   # Set to true to enable cross-run learning
  proxy_url: "http://localhost:30000"  # MetaClaw proxy URL
  skills_dir: "~/.metaclaw/skills" # Where arc-* skills are stored
  fallback_url: ""                 # Direct LLM fallback when proxy is down
  fallback_api_key: ""             # API key for fallback endpoint
  lesson_to_skill:
    enabled: true                  # Auto-convert lessons to skills
    min_severity: "warning"        # Minimum severity to convert
    max_skills_per_run: 3          # Max new skills per pipeline run
  prm:                             # Process Reward Model quality gate (optional)
    enabled: false                 # Use LLM-as-judge to score stage outputs
    model: "gpt-5.4"              # PRM judge model
    votes: 3                       # Majority vote count
    gate_stages: [5, 9, 15, 20]   # Stages to apply PRM gates

# === OpenClaw Bridge ===
openclaw_bridge:
  use_cron: false                  # Scheduled research runs
  use_message: false               # Progress notifications
  use_memory: false                # Cross-session knowledge persistence
  use_sessions_spawn: false        # Spawn parallel sub-sessions
  use_web_fetch: false             # Live web search
  use_browser: false               # Browser-based paper collection

🔭 HEP-ph Physics Mode (collider_agent)

When project.profile=hep_ph and experiment.mode=collider_agent, the pipeline routes Stage 12 through ColliderAgent (Lagrangian → FeynRules → MadGraph5 → figures via Magnus cloud) instead of the default Python ML sandbox.

Mid-stage HITL

Stage 10 (CODE_GENERATION) becomes a HITL gate. The pipeline pauses with collider_plan.md open in $EDITOR so you can review or edit the physics prompt before ColliderAgent runs. Reject sends control back to Stage 9 (EXPERIMENT_DESIGN); the hypothesis from Stage 8 stays intact.

Incremental experiments (--incremental-experiment)

To add new mass points or analyses to a completed run without redoing the heavy simulation, re-launch with --incremental-experiment and either --from-stage CODE_GENERATION (also edit the prompt) or --from-stage EXPERIMENT_RUN (reuse existing prompt):

python -m researchclaw run --profile hep_ph --output artifacts/<run_id> \
    --from-stage CODE_GENERATION --incremental-experiment

The Stage 12 sandbox will:

1. Snapshot the existing stage-12/ tree to stage-12_v{N}/.
2. Save the previous collider_plan.md as collider_plan.prev.md.
3. Build workspace_manifest.json listing reusable artifacts.
4. Send ColliderAgent a three-part prompt: CONTINUATION CONTEXT + PRIOR PLAN + your new delta.
5. Merge ColliderAgent's new results.json with the snapshot's prior one (metrics: new wins on collisions, old-only kept; artifact lists: concat + dedupe). The merge is recorded in incremental_merge.json.

Stage 13 then promotes the merged state to experiment_final/ as before.

Note: re-entering at Stage 13 alone is a no-op in collider mode and will NOT run any new physics — Stage 13 is a shutil.copy2 passthrough. PIVOT (Stage 15 decision) intentionally remains destructive because changing the hypothesis makes prior events invalid.

🙏 Acknowledgments

Inspired by:

• 🔬 AI Scientist (Sakana AI) — Automated research pioneer
• 🧠 AutoResearch (Andrej Karpathy) — End-to-end research automation
• 🌐 FARS (Analemma) — Fully Automated Research System

📄 License

MIT — see LICENSE for details.

📌 Citation

If you find AutoResearchClaw useful, please cite:

@misc{liu2026autoresearchclawselfreinforcingautonomousresearch,
      title={AutoResearchClaw: Self-Reinforcing Autonomous Research with Human-AI Collaboration},
      author={Jiaqi Liu and Shi Qiu and Mairui Li and Bingzhou Li and Haonian Ji and Siwei Han and Xinyu Ye and Peng Xia and Zihan Dong and Congyu Zhang and Letian Zhang and Guiming Chen and Haoqin Tu and Xinyu Yang and Lu Feng and Xujiang Zhao and Haifeng Chen and Jiawei Zhou and Xiao Wang and Weitong Zhang and Hongtu Zhu and Yun Li and Jieru Mei and Hongliang Fei and Jiaheng Zhang and Linjie Li and Linjun Zhang and Yuyin Zhou and Sheng Wang and Caiming Xiong and James Zou and Zeyu Zheng and Cihang Xie and Mingyu Ding and Huaxiu Yao},
      year={2026},
      eprint={2605.20025},
      archivePrefix={arXiv},
      primaryClass={cs.AI},
      url={https://arxiv.org/abs/2605.20025},
}

_{Built with 🦞 by the AutoResearchClaw team}