Menu
SkillsMP has collected 38 skills from Chaddacus/chad_claude_configs. Open a skill to review its source and details.
Showing 38 of 38 collected skills.
Wire the governed runtime into Claude Code agent teams. Classifies work into R1–R5, spawns bounded swarm teams for R3/R4, manages packet execution with reviewer barriers, enforces postflight gates, and sends completion notifications.
Your Codex companion pet — show stats, level progress, pet it, hatch it
Debug live Zoom meeting-agent failures for Rebecca/Bighead. Use when messages are missed, meeting jobs are stuck, transcripts stop, or live triage needs an event tail.
Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.
Pull Lead B (Matt Keuning) lane status for the ZW→VIP consolidation sprint from his tailnet-served endpoint, and reconcile it against authoritative repo state. Use when you need to know where Matt's lane stands (M03.x, gate-20 conditions, cross-lane triggers)…
Governs all work on the gateway consolidation project (~/gateway_consolidation). Use this skill BEFORE any implementation, testing, or investigation work on the consolidated gateway. It defines the task lifecycle, shadow comparison procedures, and the work…
Onboard a new repository into DevRelay's spine environment — push to Gitea, build the test-env Docker image, write the environmentRef, and pre-pull. Use when a repo needs to run on the fleet for the first time.
Local Qwen3.6-27B MTP on this Mac via llama.cpp — starting llama-server, MTP/sampler flags, wiring agents to the local endpoint, verifying speculative decoding, debugging load/response issues.
Run cc-cost on session transcripts: per-session cost, cache hit rate, tool-call distribution, top-expensive turns, optimization advice. Use for /cost or questions about token usage.
Generate a TLDR-style daily tech brief covering AI, developer tooling, GitHub repos, and research papers, then map the result back to the current Codex setup.
Evidence-bound web + docs research — WebSearch to discover, curl to verify, cite real URLs, tag claims by verification tier, drop what can't be grounded. Use when decisions need external grounding (market, technical, API, science). Produces a sourced research…
Manage Chad's Discord servers via REST API (bot token from Bitwarden) — channels, roles, permissions, topics, posting. Use for setup, restructure, audit, or posting to Discord.
Daily self-improvement loop — scans GitHub, arxiv, and Claude community sources for new patterns, diffs against current setup, implements Quick Wins, writes report. Use --dry-run for report only.
Self-evolving dev mode — pops the next task from the evolve queue, runs /drive --local-worker, extracts observations, auto-patches prompts/skills/presets from failure patterns, prints fitness deltas. Use when the orchestrator should improve across builds…
End-of-task composite workflow — verify (typecheck/tests/lint) → simplify → what-would-chad-do reflection → commit-push-pr. Use when implementation work is code-complete and you are about to call the task done. Not for R1 lookups or mid-implementation…
Use when a coding task should adapt to durable user/project preferences stored in omni-mem; skip for trivial factual/chat requests.
Use when the task requires automating a real browser from the terminal (navigation, form filling, snapshots, screenshots, data extraction, UI-flow debugging) via `playwright-cli` or the bundled wrapper script.
Codebase scanner — architecture mapping, security scan, pattern search across a repo. Cached index auto-loads on session start so Claude doesn't re-learn the codebase every session.
Autonomous enterprise maturity audit — scores the codebase against a 12-category rubric, fixes gaps via /build, gates on E2E tests, repeats until enterprise-ready.
Procedural commands for auto-runtime tracks, completion records, memory workflow, and route-specific work gates. Load when you need the specific commands — the behavioral rules live in CLAUDE.md.
Use when work is non-trivial and requires deterministic, fail-closed quality gates for planning and implementation evidence before final approval.
Autonomous iterative development harness. Executes a full goal without stopping to report — plan, implement, test, fix, and verify in a continuous loop. Adjusts task priorities as test results arrive. Use when you want Claude to drive implementation…
Score any codebase against an OpenShield security-maturity rubric, produce a prioritized remediation roadmap, and optionally drive fixes to completion. Invoke when the user asks for a security audit, security rubric score, AI-threat audit, or readiness check…
Send completion notifications through SMS, WhatsApp, or desktop. Use when the user asks for completion alerts, "text me when done", or "notify me" behavior for long-running tasks.
Find files with low test coverage that have changed recently. Runs pytest or jest/vitest, cross-references with `git log --since`, and emits a Markdown report ranking 'no test touches it' first, 'coverage below threshold' second.
Analyze a codebase for quality issues across code smells, clean code, separation of concerns, modularity, and API-first design, then generate a prioritized refactoring report and gated execution roadmap (characterization safety net, commit discipline, delete…
Use before implementation when non-trivial product, feature, workflow, UI, architecture, or agent-runtime work is underspecified and Claude needs to establish shared intent, constraints, module/interface impact, vertical slices, acceptance checks, and…
24/7 self-improving Claude configuration loop — continuously optimizes runtime config via automated benchmarking
Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or…
Consolidate build observations across sessions, detect patterns, promote rules, and regenerate the chad-memory.md rules file.
Install Claude Code skills into $CLAUDE_HOME/skills from a curated list or a GitHub repo path. Use when a user asks to list installable skills, install a curated skill, or install a skill from another repo (including private repos).
Scan local session transcripts to find installed skills that haven't been invoked recently. Produces a report of active vs stale skills and flags stale ones as removal candidates. Use when skill count is growing and you want to reclaim context window.
Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.
Execute Chad 2.0's autonomous build protocol for end-to-end implementation work. Use when the task is to spec, scaffold, implement, validate, and package a buildable outcome.
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or…
Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Codex's capabilities with specialized knowledge, workflows, or tool integrations.
Install skills into $CLAUDE_HOME/skills from a curated list or a GitHub repo path. Use when a user asks to list installable skills, install a curated skill, or install a skill from another repo (including private repos).
Run a final Chad reflection before closing implementation work. Use when coding work appears complete, momentum is slowing, or the agent is about to stop and needs to decide whether there is one more high-leverage, local, reversible step toward the real goal.