{"id":"aiming-lab-autoresearchclaw","name":"AutoResearchClaw","af_score":56.0,"security_score":47.5,"reliability_score":47.5,"what_it_does":"AutoResearchClaw is a Python-based autonomous research pipeline that takes a user’s research topic/idea and generates a conference-ready paper end-to-end (scoping, literature discovery/collection, synthesis, experiment design/execution in sandbox, analysis/decision loops, paper writing in LaTeX, and citation verification). It can be run via a CLI, used programmatically via a Python API, or integrated through OpenClaw/ACP-compatible agent backends, including a bridge for messaging platforms and scheduled runs.","best_when":"You want to explore and iterate quickly on research ideas using LLMs plus external literature sources and sandboxed computation, and you can accept that outcomes should be verified by humans.","avoid_when":"You require deterministic outputs, strict compliance guarantees, or you cannot provide/secure the necessary API credentials and execution/network policies for the pipeline and its dependencies.","last_evaluated":"2026-03-29T14:55:20.922114+00:00","has_mcp":false,"has_api":false,"auth_methods":["Environment variables for LLM provider API keys (e.g., OPENAI_API_KEY)","Config-based provider credentials (api_key_env + base_url)","ACP-compatible agent CLI backends (agent handles its own auth/credentials)"],"has_free_tier":false,"known_gotchas":["Highly stateful multi-stage pipelines (23 stages) may re-trigger expensive external calls unless resume/guards are correctly configured.","Uses multiple optional adapters/integrations (OpenClaw bridge, web fetch, browser option) which can change execution/network behavior and cost profile.","Delegating to ACP-compatible agent CLIs means error behavior/auth failures may surface from the external agent rather than from AutoResearchClaw itself."],"error_quality":0.0}