Browse agent skills

Git-Native Agent Memory: commit conventions (chore · refactor · spec plan/impl) enforced by a git hook, with opt-in plugins (lore, run, release, multi-agent).

Native computer-use CLI and MCP tools for existing coding agents, with browser automation and a live desktop viewer.

SOFTURE agent skills: a spec-driven delivery workflow (shape → prd → roadmap → plan → implement → review) plus parallel worktree orchestration

Talking avatars for voice agents, by InterviewFlowAI: real-time lip sync for 3D avatars in the browser, and photoreal live avatars streamed to your page. MIT licensed.

Use your Claude and ChatGPT subscriptions in Cursor, Cline, Aider, Claude Code and the Agent SDK — at subscription pricing, not per-token API bills. One local Anthropic + OpenAI-compatible endpoint: either plan answers either wire shape, with automatic fa

Agent Skill for 15–90 s motion graphics, promo and character videos: reference analysis, seven stage gates, Remotion and HyperFrames engines.

Pi footer with a compact two-line layout for model, thinking, context, cache, subscription quota, session time, Git, and Context Mode

Stripe Payments and Webhook engine for Ferrox-Node

WHYVIX Remote adalah remote development bridge untuk menghubungkan ChatGPT dengan workstation yang telah diotorisasi melalui akun WHYVIX.

Scan n8n workflow exports for OpenAI, Anthropic and Google Gemini model shutdowns, OpenAI endpoint shutdowns (Assistants API, Videos API, prompt objects, ...), and nodes removed or changed in n8n 3.0.

Grandpa makes AI coding agents work smarter, not harder. Zero-bloat architecture, 8 skills, CLI scanner, and benchmark harness for 25+ AI agents.

Crawl and extract web pages into notedown-ir with Node-API bindings and a TypeScript CLI.

Self-hosted image processing HTTP API

Agent skills for Pandark users. Crawl and extract web pages with @notedge/pandark. Install with npx @notedge/pandark-skills.

Open source AI code review for Claude Code and Codex, before you push. Scanners (SAST, secrets, dependencies, lint) on the lines you changed, then a separate reviewer process that checks every scanner finding and is given every changed line. No other API key.

Public token, state-attribute and capability contract shared by the Whatalo platform and every theme

Question-driven decision trees for planning software features: JSON in each app's .decisions/, an agent CLI, and a local HTML viewer for human-agent coordination. Ships as an agent skill.

Write a team playbook once, run it in every AI assistant: Claude Code, Codex, GitHub Copilot, Cursor, Gemini CLI, ChatGPT and Microsoft 365 Copilot. No server, no API key.

Database explorer for DeepSeek Harness: connect SQLite / MySQL / PostgreSQL / SQL Server / MongoDB / ClickHouse / Redis / Elasticsearch / Qdrant (vector DB) — zero-dependency NoSQL & vector engines, browse schemas and run SQL — a PyCharm-Database-style pa

Configurable subagent rosters for Pi: write a profile, then run work as an observable pipeline

由 AI 自动完成 AutoDL 实例开关机的 Claude Code skill，含空闲自动关机、预算核对与用量记账

gate-hub integration kit: token and webhook verification, OIDC relying party, directory mirror, revocation guard, Fastify/Express/tRPC adapters, Postgres stores and a testing kit

Telegram Bot API client — reusable across projects

The Graview command line: start a product, check its declaration, write its agent docs, serve its store, preview a document with no product and photograph its places, say whether a product is on one release, host it for an agent over MCP, apply an act to

FFmpeg video encoding and audio mixing pipeline for OBSIDIAN FLOW

Job postings API client: live US jobs verified on employers' own career sites, still-open checks, full feed, and a competitive hiring report for one healthcare, finance, IT, engineering, science, HR, legal or marketing role.

Offline-first contextual discovery of curated AI coding skills and plugins

OS keychain credentials and local outbound privacy guardrails for Pi: confirm, obfuscate or cancel sensitive data before a model request.

Run an isolated QA agent in PI Agent, capture auditable execution evidence, and evaluate explicit targets with the selected Pi model.

Delegate separable work to autonomous subagents that run beside a PI Agent session and report back with evidence.