Scout a codebase to identify high-value work, plan the fixes (point and architectural), and produce a ranked map from codebase to tasks -- each tagged with a recommended model -- without ever implementing the fix. Cheap-model agents sweep and score every file by Impact x Opportunity (1-5); the router then runs meta-level pattern analysis, plans point and architectural fixes, and recommends a model per task. Team: target guard, skip gate, formula, profile, parallel scorers, ranker, pattern-grouper (meta-level analysis + architectural fix planning), task-writer (point-fix planning), model-recommender. Objectives: bugs, security, tech debt, dead code, performance, conversion.
Live savings status line + terminal dashboard, and /distil · /distil-stats · /distil-shadow · /distil-dashboard commands for distil context compression
Generate arc42 architecture documentation from a codebase. Scans the repo into a structured evidence base, then writes the 12 arc42 sections with Mermaid diagrams where the code gives high-confidence structure and explicit GAP flags where human input is needed (never fabricated). Knowledge base derived from arc42 under CC BY-SA 4.0.
Critique, improve, and create Claude Code skills against one standard: a skill is a correction you got tired of making. Target the step that fails, use leading words that recruit training priors instead of coined terms, keep skills small and user-invoked, grow them only from observed corrections, and delete every instruction that does no work. critique reports findings with fixes and edits nothing; improve prunes no-ops, swaps explanation for leading words, and folds recurring corrections from session history in as one-sentence fixes; create mines session history for repeated corrections, checks overlap and default behavior before writing, and drafts a minimal skill. Subcommands: critique, improve, create.
Complete development environment setup for IOWarp Core — Docker devcontainers, builds, debugging, and IDE configuration. Gets any team member from zero to a running dev environment.
Build, test, and debug event-driven state machines in Laravel. Includes full VitePress documentation (87 pages) and agent-optimized cheat-sheets for testing, delegation, parallel states, and QA setup.
A skill that teaches coding agents to write correct C# with the WebDriverBiDi.NET client for the W3C WebDriver BiDi protocol: connecting, commands, events, scripts, and errors.
Build, test and polish 2D games with Ludra, a game engine designed for agents: JSON scenes, fixes for every error, headless play-testing, SVG art, particles and sound.
if you want to go fast, go deep first. pstack helps you write less, but higher quality code. rigorous agent workflows you can parallelize with confidence. Third-party plugin by Lauren Tan, ported from Cursor to Claude Code.
Six-skill audit suite for iOS/Swift codebases. Every finding is classified on a 3-axis framework (real bug / scatter / dead code) with mandatory coaching citations to patterns in the audited codebase.
The official Jadis skill: the model, components, templates, communication, routing, translations, server rendering, tests, the mistakes other frameworks' habits cause, and a self-check.
Constraints, alternatives already ruled out, and warnings left by whoever was here last — injected before the agent edits the path they cover. Records are ordinary git trailers: they survive rebase and rename, and squash keeps them only when `commitlore squash-preserve` or its GitHub Action runs.
Capture any live web app: demo videos, screenshot sets, and scripts; a full UI component library with DTCG design tokens, a Claude Design handoff zip, and a shadcn/ui map; visual and code inconsistency audits.