skills/researcher/SKILL.md
Use when analyzing repositories, conducting deep research on codebases, performing architecture reviews, or exploring large projects from a git URL.
npx skillsauth add tao3k/omni-dev-fusion researcherInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Sharded deep research for large repositories using the Qianji runtime (xiuxian-qianji) and a suspend/resume approval loop.
┌─────────────┐ ┌────────────────┐ ┌──────────────────┐
│ Setup │ --> │ Architect Plan │ --> │ Await Approval │
│ clone + map │ │ shard proposal │ │ suspend/resume │
└─────────────┘ └────────────────┘ └──────────────────┘
│
▼
┌──────────────────┐
│ Deep Analysis │
│ approved shards │
└──────────────────┘
git_repo_analyerCore command to execute repository research via Qianji.
Parameters:
repo_url (string, required): Git repository URL to analyze.request (string, optional): Research goal. Default: "Analyze the architecture".action (string, optional): "start" or "approve". Default: "start".session_id (string, required for approve): Session returned by start.approved_shards (string, required for approve): Approved plan JSON string.Execution model:
action="start":
session_id, proposed_plan, and approval prompt.action="approve":
The command returns structured JSON. Typical fields:
successsession_idmessage / proposed_plan (start phase)analysis_result / full_context (approve phase)xiuxian-qianji (Rust).workflows/repo_analyzer.toml.researcher/
├── SKILL.md
├── README.md
├── references/
│ └── run_research_graph.md
├── workflows/
│ └── repo_analyzer.toml
tools
Internal agenda skill background.
tools
Agenda skill
tools
Agenda skill
tools
Agenda resolver fixture.