keel-research
Skill0 starsChanged 28 days agoMITView on GitHub
Research lane router. Routes knowledge tasks — querying your own documents, real-time web search, and building a searchable second brain — to the right vendored skill (notebooklm-skill, ai-second-brain) or MCP (perplexity). Defers to keel-v2. Use for grounded research, citation-backed answers from your notes, live web lookups, or mining your own AI history.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- muratsilahtaroglu/claude-code-starter-kittheir own version, first seen Sep 25, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- muratsilahtaroglu/claude-code-starter-kit★ 9same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository16 other files it gives its agents.
- skills/keel-v2/CLAUDE.mdCLAUDE.md
- doc-syncSkill
- fused-landingSkill
- keel-contentSkill
- keel-designSkill
- keel-devSkill
- keel-financeSkill
- keel-legalSkill
- keel-marketingSkill
- keel-opsSkill
- keel-qaSkill
- keel-secureSkill
- keel-setupSkill
- keel-v2Skill
- keel-verifySkill
- lazy-loopSkill
The file
Files
Keel Research — research lane router
Index, not an implementation. Routes the research task; the specialist runs.
Deference — keel-v2 is the arbiter
Research/answer tasks that produce no diff are the keel-v2 no-artifact lane — answer directly, no plan/branch/tripwires (keel-v2 §0). Rigor only re-enters when research output becomes a committed artifact.
Dispatch table
| When the task is… | Dispatch to | Type | Source |
| Query your own Google NotebookLM notebooks, citation-backed answers | notebooklm-skill | skill | PleasePrompto/notebooklm-skill |
| Real-time web search / deep research via Sonar | perplexity | MCP | perplexityai/modelcontextprotocol |
| Build/search a Karpathy-style wiki from your ChatGPT/Claude history | ai-second-brain | skill | charlie947/ai-second-brain |
Precedence & overlap
- Grounded in your docs →
notebooklm-skill. Grounded in the live web →perplexity. Grounded in your past conversations →ai-second-brain. Pick by source-of-truth, not by habit. - The harness also ships a built-in
deep-researchskill (multi-source, adversarially verified, cited report). For a big multi-source report prefer it; useperplexityfor quick real-time lookups.
If the target isn't installed
Say so; fall back to the built-in deep-research skill or base web tools.
Install: ./install.sh research.
Sources: PleasePrompto/notebooklm-skill · perplexityai/modelcontextprotocol (MCP) · charlie947/ai-second-brain. Upstreams unmodified.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.