Dev & Engineering Claude Skills
Effect JSDoc Authoring Skill
Write, insert, or fix Effect public API JSDoc so it passes the jsdocs oxlint rule.
Android Adaptive UI Skill
Make Jetpack Compose apps adaptive to phones, tablets, foldables, desktops, and more — with multi-pane layouts and flexible components.
Skill Creator
A guided skill-authoring workflow that helps you build well-structured, context-efficient custom Skills for AI agents, with init and packaging scripts included.
Maintainer Response (agent-service-toolkit)
Drafts GitHub issue and PR replies for the agent-service-toolkit repo in maintainer JoshuaC215's authentic voice — always drafts for human review, never auto-posting.
Engram Persistent Memory Protocol
Persistent memory for AI coding agents—auto-save decisions and discoveries across sessions and compactions.
Loki Checkpoint Mode
Makes an autonomous coding agent pause every N tasks, generate a summary, and wait for human approval — balancing autonomy with control.
TradingView Lightweight Charts Skill
Teaches AI coding assistants the real Lightweight Charts v5 API conventions so generated chart code avoids common time, marker, and plugin foot-guns.
Playwright Browser Automation Skill
Lets coding agents write and run real Playwright scripts on the fly for page testing, screenshots, login flows, and arbitrary browser automation.
Node Modules Inspector — Dependency Audit Skill
Inspects your project's installed node_modules to surface duplicated packages, disk-hog dependencies, and outdated dependency declarations ripe for upgrade.
BlockWatch
Link co-dependent code with inline <block> tags so docs and configs can't silently drift — violations fail CI or pre-commit.
ModLens — Plug-in Vision for Text-Only Models
Gives text-only coding models sight: paste an image, get structured JSON evidence — OCR, layout, and semantics — via a vision bridge.
Xberg Format-Specific Extraction Workflows
A reference skill for agents that spells out the exact extraction paths and security validation steps Xberg uses for DOCX, PDF, archives, structured text, and email formats.
Desktop Commander MCP
Give your AI agent full control of your computer: persistent shells, background processes, filesystem reach, structured documents, and large-scale search.
OpenMAIC Multi-Agent Interactive Classroom Skill
A guided skill that deploys OpenMAIC and turns any topic or document into an interactive classroom with AI teachers and AI classmates.
Remote GPU Trainer Skill
One Agent Skill for the whole DL experiment lifecycle: run a job (RUN), verify the number is real (VERIFY), and deliver reproducible single-source figures (DELIVER) — on a GPU you own or rent.
Swift Concurrency Pro
Helps AI coding assistants write correct Swift concurrency code, targeting the async/await, actor, and structured-concurrency mistakes LLMs actually make.
peon-ping Config Manager
Easily tweak peon-ping's volume, pack rotation, categories, and more by editing its config file.
PinMe Worker OpenRouter API Integration
Call OpenRouter-backed LLM APIs from a PinMe Worker (TypeScript) with streaming and web search support.
SkillOpt-Sleep: Sleep-Cycle Self-Improvement for Your Local Claude Agent
Give your Claude agent a nightly offline sleep cycle that reviews real sessions and keeps only validation-gated improvements — no model-weight training.
SkillOpt-Sleep
Gives your local Codex agent a 'sleep cycle': offline review of past sessions that keeps only validation-gated improvements — no model-weight training involved.
Optional-Dependency Smoke Tests (agent-service-toolkit)
When your change touches the Postgres/MongoDB checkpointers, the AG-UI endpoint, or LangFuse tracing, run real dependencies in Docker and verify the integration end-to-end without waiting for full CI.
SkillCheck Free
A read-only linter skill that makes your AI assistant validate SKILL.md files against the agentskills spec, catching structural, semantic and security issues before your users do.
cmux Workspace Skill
Lets an AI coding agent work safely inside the cmux workspace that invoked it, without disturbing the workspace or window the user is actually looking at.
e2e Verify — End-to-End Change Verification Skill
Prove every change with the real CLI against the testbed and benchmark apps — visible evidence, not "it compiles".