Browse agent skills

类型/接口/能力声明包（只导出 `type` 与 `*Capabilities`，不含运行时逻辑）

DeepSeek LLM 提供者（深度思考 + 工具调用）

Turn Claude Code into its own Meta-Harness — a skill that evolves the scaffolding around a fixed model (memory, retrieval, context, prompts) via a native propose→score→Pareto loop. Native reimplementation of Meta-Harness (Lee et al. 2026).

Firecrawl web scraping tools for Pi coding agent - scrape, crawl, map, and search with Firecrawl API v2

Static due-diligence scanner for local AI model artefacts and repositories.

Native LangChain chat models for OpenAI Codex subscription OAuth and OpenCode Zen/Go — no Node sidecar.

DAIHUM tool authority: the transport-agnostic contract for assistant/agent tool calling — authority port, 4-state invoke response, 6-axis authority facts, and the pure effect-class → facts policy. Zero dependencies. Elevated from carrel ops-protocol.

Framework-agnostic core for using Claude Code's OAuth subscription as an Anthropic provider: credential read/refresh, signed billing header, beta computation, and request transforms. No Pi, no LangChain — just the native Claude Code stack.

Context-Driven Development for AI coding agents — install Draft into Claude Code, Cursor, Codex, or opencode.

A Claude skill that builds a great GitHub profile README — 9 archetypes distilled from 181 real profiles, with templates and a widget generator.

n8n community node for Infoblox Universal DDI and NIOS WAPI

Agent skill that makes Claude Code, opencode, and other SKILL.md agents write and review code to Clean Code / Clean Coder standards.

Agent2Agent (A2A) server adapter backed by an Agent Host Protocol client.

Open-source, self-hosted, OpenAI-compatible LLM gateway — multi-provider routing, fallback, semantic cache, budget caps, PII guardrails, audit logs

Agent Host Protocol gRPC transport bindings.

Audits time-based machine learning backtests, walk-forward validation, rolling retraining, forecasting evaluations, and temporal train/test pipelines for data leakage, faulty split logic, invalid feature timing, target leakage, calibration/tuning leakage, and misleading experiment comparisons.

Multi-agent smart contract security auditing framework for Solidity (EVM) and Rust (Solana/Anchor) — installable via npx

Production-shaped TypeScript Agent Host Protocol server with pluggable agent adapters.

Core skill for the pm-tasks-* family: extracts implementation plans into structured task content, defines the CRUD vocabulary (task.create, checklist.check, task.close, task.due-date.set, task.assignee.add, task.comment.add), specifies the autonomous-mode

Trello adapter for the @llodev/pm-tasks-* family. Use when the user mentions Trello (create card, publish to Trello, post to Trello, --publish, check checklist item, close card, comment on card) or wants to publish a plan as Trello cards. Modes: paste-rea

Project-local AI video production kit for Remotion, HyperFrames, Codex, Claude, Gemini, Ollama and OpenRouter workflows

LLM usage telemetry SDK for Cloptima reporting, attribution, and analytics

Failure-first spec stress test — inverts sycophancy to surface failure modes (C2/P1).

Permissions and environment preflight check before fan-out (D2 phase 0).

Shared core for ADLC tools: LLM providers, git helpers, CLI conventions, ledgers, tickets, mutations.

rag-reviewer is a testing & review skill indexed by Skillful.sh. Package: rag-reviewer.

An open source local-first terminal coding agent. Peer programming with any frontier model or local/free models to run agentic development.

Install the signed Infinite AI marketing Desktop app for macOS.

Google Antigravity OAuth auth plugin for OpenCode - access Gemini 3 and Claude 4.6 using Google credentials

CLI tool for ReportForge — render PDFs from the command line