
from researchclaw-skill8
Chinese-language wrapper for AutoResearchClaw: a 23-stage autonomous research pipeline that automates literature retrieval, experiments, drafting, and citation
ResearchClaw (Chinese) is a localisation and orchestration wrapper around AutoResearchClaw that exposes a 23-stage autonomous research pipeline in Chinese. Given a research topic it helps set up prerequisites, run staged experiments (simulated or sandboxed), gather and analyze results, generate paper drafts, run multi-agent peer review, and verify citations. The skill provides CLI helpers for setup, configuration, run monitoring, diagnosis, and recovery — and is explicit about what it cannot do (e.g., start system Docker daemon or provide LLM API keys).
Use this skill when you want an agent to run an end-to-end research workflow in Chinese: setting up environment dependencies, interactively creating a config.yaml, executing the 23-stage pipeline, monitoring progress, diagnosing failures, and exporting papers. It supports simulated runs for quick iteration or sandbox/remote execution for real experiments.
/researchclaw:setup, /researchclaw:config, /researchclaw:run, /researchclaw:status, /researchclaw:diagnose, /researchclaw:resume, /researchclaw:validate) and explicit guidance on validation, recovery, and common failure modes.This skill targets coding-capable agents and CLI-execution environments (Claude Code, Copilot-like agents, Cursor, Codex, and other agents able to run bash/python helpers). It assumes Python 3.11+, Docker and LaTeX availability for full functionality.
This skill has not been reviewed by our automated audit pipeline yet.