efitz-skills

Dependency discovery toolkit: enumerate what a system depends on across six categories — packages, out-of-project services, configuration, secrets and permissions, OS and cloud resources, and network paths — each with the failure-relevant declarations (timeout, retry, fallback, health check) found for it. Static and read-only; emits a versioned JSON contract consumed by downstream analysis.

8 skills

Localization/i18n toolkit: analyze missing keys, measure coverage, detect non-localizable strings, translate, validate translations, update JSON locale files, and backfill every locale. Use for i18n auditing, translation, and locale-file maintenance.

7 skills

Portable CATS API fuzzing toolkit: bootstrap per-repo config (init), run fuzz campaigns through config-declared shell hooks (run), query and render results from SQLite (report), triage true positives into a remediation plan (analyze), and manage declarative false-positive rules (fp). Repo-specific setup lives in .local/cats/config.yaml; false-positive rules live in a committed YAML file.

5 skills · 1 agent

Developer toolkit (sem-powered): annotate code with durable SEM@<sha> intent markers (sem-annotate), keep them maintained via a project convention (sem-auto), and find dead code + duplication and produce a ranked plan (dedupe). Uses the sem CLI entity graph. Supports Go, TypeScript/JavaScript, and Python.

3 skills · 3 agents

Logseq graph toolkit for a local classic (file-based) graph: capture notes/TODOs into journals or pages (capture), answer questions from the graph (query), find consistency problems (lint), merge/dedupe and restructure pages (organize), and import notes from an Obsidian vault (from-obsidian). Hybrid file/HTTP-API access with automatic graph discovery.

5 skills

Project discovery toolkit: identify what a codebase is built with (stack), what its requirements documents, PRDs, specs, and wiki pages say it must do (docs), how it deploys and what it depends on (topology), and what workflows its users actually perform (journeys). Read-only inference backed by a deterministic inventory script; emits versioned JSON contracts consumed by other plugins.

5 skills

Integration test design toolkit: discover a project's test conventions and how integration tests are separated (conventions), assess the quality of existing tests for over-mocking and misleading coverage (critique), find state-establishment and teardown affordances (state), and synthesize a prioritized scenario plan from confirmed customer journeys and documented requirements (design). Reports conflicts between what the documentation requires and what the code does. Read-only discovery; requires the profile plugin.

4 skills

GitHub workflow toolkit: pick the next issue to work on from a milestone/backlog (backlog) and file detailed issues into a repo/Project using a locally-provisioned metadata cache (create-issue). Use for issue triage and issue filing.

2 skills · hooks

Session continuity toolkit: write a machine-local HANDOFF.md holding only current state (handoff) and resume from it by checking it against live repo facts, reporting drift, and stopping before any work starts (go). A SessionStart hook reminds Claude to run go when a HANDOFF.md exists.

2 skills · hooks

OpenAPI workflow toolkit: locate and record the project's OpenAPI and Arazzo specs (init), and generate an Arazzo 1.0 workflow spec from the confirmed customer-journeys doc mapped onto real operations (arazzo). Prose-driven with human gates; validators used when installed.

2 skills

Security toolkit: vet plugins/skills before install (vet-plugin) and audit code for race conditions and concurrency bugs (race-cond). Use before installing third-party agent extensions or when reviewing concurrent code.

2 skills

Update dependencies safely across Go, Python, and Node ecosystems. Use when the user asks to bump, update, or upgrade dependencies, packages, or deps; run a dependency bump/upgrade; fix Dependabot or security advisories; or refresh outdated packages. Triggers on phrasings like 'bump the deps', 'bump dependencies on <branch>', 'update packages on main', or 'run a dep upgrade'. Auto-detects ecosystems (Go/Python/Node), applies safe patch and minor updates with build, test, and lint validation, bisects failures to isolate bad packages, and surfaces a prioritized plan for major or held packages that need manual review.

1 skill

Environment readiness toolkit: checks whether the CLI tools, config files, and auth sessions the other plugins depend on are present and working (check). Reports what's missing, why it matters, and how to fix it; report-only by default, with an explicit --fix mode that confirms and installs declared remedies for the detected platform.

1 skill

UI testing toolkit: triage Playwright visual-regression (screenshot) failures — baseline vs actual vs diff, framed against task context, to decide bug vs expected change (vrt).

1 skill

Documentation toolkit: verify a doc's accuracy against source code and external references, then migrate it into a project wiki (verify-doc).

1 skill

Evaluate technical business writing for "boringness" across 20 sub-dimensions on four axes (Direction, Density, Texture, Surprise). Combines a mechanical analyzer (15 sub-dimensions, deterministic, runs as a Python script) with five LLM-judged sub-dimensions that require semantic judgment. Use when the user asks to review a document for engagement, clarity, or "is this boring", or for prose-mechanics issues in technical writing.

1 skill