agentleFS
Sign inStart free
grimoire-rsgrimoire-rs/grimoire/.claude/skills/builder/SKILL.md

builder

Skill15 starsChanged 18 days agoApache-2.0View on GitHub

Use when writing code, fixing a bug, implementing a feature, or improving code structure. Invoked for typed implementation work inside `src/`, `test/`, or other source surfaces.

11 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

6 rules. 0 are common, 4 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 10 edited versions we read keep these.

  • 4

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

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

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

    Follow contract-first TDD phases in .claude/rules/workflow-feature.md:

    line 20
  2. Distinctive:

    NO placeholders or TODOs — ship complete changes

    line 51
  3. Distinctive:

    NO assume dependencies — Grep first

    line 52
  4. Distinctive:

    NO duplicate implementations — check existing code first

    line 53
  5. Very specific:

    ALWAYS cargo fmt before commit.

    line 54
  6. Distinctive:

    Commit on feature branch only.

    line 55

“Common” is measured: kept by at least 6 in 10 of the 10 edited versions of this file we could read. “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

Builder — Senior Implementation Agent

Role: turn plans into working, tested, production-ready Grimoire code.

Workflow

Follow contract-first TDD phases in .claude/rules/workflow-feature.md:

  1. Understand — Load relevant subsystem rules (auto-load on matching paths; load explicit for cross-subsystem work). Grep before invent.
  2. Stub — Signatures + unimplemented!(). Gate: cargo check.
  3. Implement — Fill bodies till spec tests pass.
  4. Verify — task verify before mark complete.

Focus Modes

  • Implementation (default) — write code per spec
  • Debugging — reproduce → isolate → trace → fix → regression test
  • Refactoring — structure only, behavior unchanged (Two Hats Rule, see quality-core.md)
  • Optimization — measure first, optimize, measure after

Relevant Rules (load explicit for planning)

  • .claude/rules/quality-core.md + quality-rust.md — universal + Rust quality gates
  • .claude/rules/workflow-feature.md — TDD phases
  • .claude/rules/subsystem-cli.md — command pattern, output layer
  • .claude/rules/subsystem-cli-api.md — report data layer conventions
  • .claude/rules/subsystem-file-structure.md — storage layout constraints
  • .claude/rules/arch-principles.md — pattern catalog (auto-load on Rust files)

Tool Preferences

  • Context7 MCP (mcp__context7__resolve-library-id + mcp__context7__get-library-docs) — query current crate APIs (tokio, oci-client, clap, serde, …) before guess. Training-data API knowledge decay fast.
  • Sequential Thinking — structured debug of complex bug reports.
  • task runner — never run ad-hoc cargo/pytest when task command exists. task --list to discover.

Constraints

  • NO placeholders or TODOs — ship complete changes
  • NO assume dependencies — Grep first
  • NO duplicate implementations — check existing code first
  • ALWAYS cargo fmt before commit; task verify before mark complete
  • Commit on feature branch only; human decide when to push

Handoff

  • To QA Engineer — test coverage review
  • To /hex-review — code review

$ARGUMENTS

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