agentleFS
Sign inStart free
r-irber-irbe/proof-skills/skills/lean-package-research/SKILL.md

lean-package-research

Skill3 starsChanged 24 days agoApache-2.0View on GitHub

USE FOR: Lean 4 package and toolchain research: evaluating Reservoir/GitHub packages, Mathlib/cslib pin changes. DO NOT USE FOR: ordinary theorem lookup without a package decision (@lean-research). TRIGGERS: Lean package, Reservoir, GitHub Lean repo, Lake dependency.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    STOP if the package would require editing lean-toolchain, lakefile.lean, or lake-manifest.json without an explicit edit claim and human …

    line 47
  2. Distinctive:

    STOP if the adoption class is ADOPT-NOW but confidence is below the local belief floor, validation is unknown, or the dependency graph is …

    line 49
  3. Distinctive:

    STOP if a package is archived, stale, or has unclear transitive dependencies and the caller asks for immediate adoption.

    line 51

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

lean-package-research

Package research is separate from theorem lookup because dependency decisions change reproducibility, build time, trust surface, and future maintenance. This skill turns a candidate package or pin change into an adoption recommendation with evidence, sequencing, and validation.

Routing

  • USE FOR: Lean 4 package and toolchain research — evaluating Reservoir/GitHub packages, Mathlib/cslib pin changes, Lake dependency health, package adoption classes, update sequencing, and external theorem-substrate candidates.
  • DO NOT USE FOR: ordinary theorem lookup without a package decision (use @lean-research); package-file mutation or lake update execution (delegate validation to @lean-enforcement under an edit claim); proof writing (use @lean-proof); general project QA (use @lean-quality-engine).
  • TRIGGERS: Lean package, Reservoir, GitHub Lean repo, Lake dependency,

Workflow

  1. Scope the package question [discover] — identify whether the caller wants theorem borrowing, dependency adoption, pin update, fork/vendor choice, or negative-result evidence. If the request only needs symbol lookup, hand off to @lean-research.
  2. Run the discovery ladder [discover] — use ../../references/discovery-ladder.md for Reservoir, GitHub, current-pin package source, theorem-search services, and literature checks. Record negative results rather than relying on memory.
  3. Assess dependency health [validate] — inspect Lake metadata, toolchain compatibility, maintenance status, archive/staleness, license visibility, transitive dependencies, binary fetches, and whether the package is already

Recovery & STOP

  • STOP if the package would require editing lean-toolchain, lakefile.lean, or lake-manifest.json without an explicit edit claim and human authorization.
  • STOP if the adoption class is ADOPT-NOW but confidence is below the local belief floor, validation is unknown, or the dependency graph is dirty.
  • STOP if a package is archived, stale, or has unclear transitive dependencies and the caller asks for immediate adoption; return RESEARCH-MORE or DO-NOT-ADOPT-NOW with evidence.

Handoffs

  • Predecessors: @lean-research for symbol/literature leads, @lean-integration-protocol for project integration boundaries, agent:gateway for package-focused user requests.
  • Successors: @lean-research for deeper theorem lookup, @lean-enforcement for validation commands, @lean-quality-engine for milestone/package-health scoring.
  • References: discovery ladder (../../references/discovery-ladder.md) and eval prompts

Detailed reference

Full encyclopedic reference, decision trees, and lookup tables live in references/lean-package-research-handbook.md .

See also

  • ../../references/lean-package-research-handbook.md -- Complete lean-package-research Handbook
  • ../lean-gateway/SKILL.md -- Top-level skill router

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.

Check your setup, free