RepoPilot

mattpocock / skills

Search mattpocock/skills

GitHub

Code

Understand mattpocock/skills's source layout and find a starting point for your task using the saved repository evidence.

Detailed import-graph metrics were not captured for this report.

How the code fits together

AI
  • Documentation & MetadataTop-level guidance, changelog, and glossary for contributors and users.
  • Engineering SkillsPrimary collection of composable agent skills organized by engineering discipline.
  • Skill Agent ConfigurationsOpenAI-compatible agent configurations for each skill enabling model execution.
  • In-Progress & DeprecatedExperimental and legacy skills kept for reference and iteration.
  • Build & ReleaseVersioning and publishing workflow configuration.

Files at 49dd158d1076 · descriptions are AI-written.

The descriptions and the file list come from different revisions. Check responsibilities against source.

Where to start reading

  3. Suggested starting pointAI
  4. Suggested starting pointAI
  5. Suggested starting pointAI
  6. Suggested starting pointAI

Find files for a task

AI

Add a New Engineering Skill

Suggested by the saved analysis. Verify the proposed change against these source files.

  • Create skill directory under skills/engineering/{skill-name}/
  • Write SKILL.md with task description, constraints, and expected outputs
  • Create agents/openai.yaml with model instructions and tool bindings
  • Add supporting documentation (ADR-FORMAT.md, DEEPENING.md) if the skill benefits from structured guidance

Create Structured Guidance for a Skill

Suggested by the saved analysis. Verify the proposed change against these source files.

  • Define domain glossary format in a GLOSSARY-FORMAT.md file within the skill directory
  • Document decision templates in ADR-FORMAT.md for architecture/design skills
  • Define iterative patterns like DESIGN-IT-TWICE.md for skills benefiting from multiple passes

Release a New Skill Version

Suggested by the saved analysis. Verify the proposed change against these source files.

  • Run npm changeset to document changes using GitHub changelog format
  • Verify version syncing with node scripts/sync-plugin-version.mjs
  • Update CHANGELOG.md to reflect new skill additions or changes

Change build or automation

Inspect the relevant configuration and the commands it runs before changing the workflow. These suggestions use file names, not verified execution paths.

  • File-name match for this task.
  • File-name match for this task.
  • File-name match for this task.

File paths come from a snapshot of up to 500 captured paths; an omitted file may still exist.

  • No public or runtime entry point can be identified from the captured paths.
  • The file list was refreshed after the written analysis. Subsystem descriptions and suggested changes may describe an older revision.

Architecture and request flow

Start here — files every contributor must understand

AI · unverified
1
README.md

Entry point documenting the repository's purpose as a collection of composable agent skills for engineering tasks, not process-owning frameworks.

2
AGENTS.md

Core documentation explaining how agents are structured and configured across the skills directory.

3
skills/engineering/README.md

Defines the engineering skills category structure and how individual skills are organized and composed.

4
package.json

Declares project metadata, versioning via changesets, and npm package manager requirements for consistency.

5
GLOSSARY.md

Defines domain terminology and concepts referenced across skill descriptions and agent configurations.

6
skills/engineering/codebase-design/SKILL.md

Example skill definition showing the standard template and expected structure for composable engineering tasks.

Request flow

Technology choices

YAML agent configurations

Language-agnostic, human-readable format for defining LLM prompts and tool bindings across multiple models.

Markdown skill definitions

Version-controllable, composable documentation that works in any editor and integrates naturally with GitHub.

Changesets + npm versioning

Decoupled change documentation from code, enabling semantic versioning and automated changelog generation.

Build time and runtime

Build-time

  • ▸
    Changesets CLI — Validates and applies semantic version updates based on changeset files.
  • ▸
    Sync plugin version script — Ensures version consistency across package.json and distributed skill artifacts.

Runtime

  • ▸
    Agent executor (external) — Interprets agent configurations and invokes LLM with skill prompts.
  • ▸
    Skill resolution — Maps user requests to appropriate SKILL.md and agent config based on directory structure.
  • ▸
    LLM model instance — Executes prompts defined in openai.yaml configurations and returns structured output.

Design trade-offs

Skills are small, composable, and adaptable rather than all-encompassing process frameworks.

Provides flexibility and avoids lock-in to a single methodology that may not fit all teams.

Users must integrate skills themselves; there is no guardrail preventing misuse or partial adoption.

Agent configurations are model-specific (openai.yaml) rather than model-agnostic DSL.

Allows exploitation of model-specific capabilities and prompt engineering best practices.

Adding a new model requires duplicating configurations; drift between implementations is possible.

Support files (ADR-FORMAT.md, DEEPENING.md) are optional per skill.

Reduces friction for simple skills while allowing complex ones to grow structured guidance.

Inconsistent user experience and variable quality of guidance across the skill set.

Non-goals

  • ✗Does not enforce a mandatory engineering process; skills can be used à la carte.
  • ✗Does not provide real-time collaboration or multi-user workflow management.
  • ✗Does not include built-in testing frameworks or continuous integration runners.
  • ✗Does not guarantee model-agnostic execution; OpenAI configurations are primary.

Code quality and hotspots

Quality Score: 75/100

Low Risk

This repo is in good health with CI, tests, and active maintenance.

No tests detected -15Multiple anti-patterns -10

Configuration status

CI/CD Pipeline — configured

Provider: GitHub Actions

3 GitHub Actions workflows

Test Suite — not detected

Add unit tests for critical paths

Test Coverage — ~0%heuristic

0 test files / 2 source files. This is a file-count ratio, not real line coverage — set up Codecov for authoritative numbers.

Dependency Scanning — no config detected

No Dependabot / Renovate / Snyk / Trivy / CodeQL config found. Enable Dependabot or Renovate.

AI-suggested anti-patterns

AI · unverified

Pattern-matches from the file structure and dependencies — useful as hypotheses, but not grounded in actual source. For real findings see Code Scanning above.

Implicit skill configuration coupling

Medium

YAML agent configs tightly couple prompts to model assumptions with no explicit fallback for model versions or capability mismatches.

skills/engineering/*/agents/openai.yaml

Optional guidance fragmentation

Low

Some skills include rich supporting docs (DEEPENING.md, ADR-FORMAT.md) while others contain only SKILL.md, causing inconsistent user experience.

skills/engineering/

Deprecated skill coexistence

Medium

Retired and experimental skills remain discoverable, risking user confusion or adoption of unstable practices.

skills/deprecated/, skills/in-progress/

AI-suggested bottlenecks

AI · unverified

Speculation from file structure, not profiling data. Validate against real metrics before acting on these.

Architectural Bottleneck

Each new model or API provider requires duplication of all skill configs; no shared abstraction reduces maintainability.

Agent configuration format (openai.yaml)

Performance Bottleneck

Linear filesystem traversal with convention-based naming; no indexing or caching for rapid skill lookup.

Skill discovery mechanism (directory structure + naming)

Operational Bottleneck

Manual changeset creation required per feature; no automation to reduce friction or enforce discipline.

Changeset-driven release workflow

Code metrics

Static analysis on sampled source files (up to 15 files).

Avg Cyclomatic Complexity0

Counts if/for/while/switch/catch/ternary branches

Errors / Warnings0 / 0

Secrets, empty catches, console.log, var usage, long lines