Deterministic workflow and safety policy enforcement for OpenCode
Directory
Browse agent skills
Search by intent: design systems, testing, deployment, security, documentation, Azure, Supabase, React, or any workflow you want an agent to perform better.
Showing 13,891–13,920 of 126,802 skills
Page 464 of 4,227 · loading 30 at a time
Name reserved for an upcoming spec-driven, multi-agent AI development framework. Not yet published for use.
Convierte videos en artículos de blog en Markdown mediante IA
A lightweight, file-based framework for building Discord bots with discord.js and Bun.
Universal provider proxy for OpenAI Codex & Claude Code — use any LLM with Codex CLI/App/SDK and Claude Code
Lint LLM chat-completion prompts for cross-provider portability issues
Docker sandbox provider for the TAI ecosystem — per-session containers on a remote engine over the Docker Engine API.
Direct/host sandbox provider for the TAI ecosystem — runs a session's code as a plain host subprocess, no container and no isolation.
Mindmap plugin for DeepSeek Harness: a plain markdown file in the working directory IS the mindmap; the chat edits it step by step and the right-side panel follows live.
Token usage statistics for DeepSeek Harness (7/30-day per-model charts + yearly heatmap) plus an inline coding-plan balance/quota readout in the composer (minimax/deepseek/kimi/openrouter/zhipu/mimo; mimo falls back to local usage since MiMo exposes no AP
Python SDK for the AI Sandbox Pilot platform (Section 8.3 / Build Order 41)
CLI for Alora, the cloud where your agent works — deploy, preview and share small apps; born private, sleep for free.
A Claude Code skill and hook that replace the built-in Artifact tool with local, version-controlled HTML decision briefs.
Unofficial SingaPay payment gateway SDK — framework-agnostic, server-only, zero runtime dependencies.
Universal context continuity & handoff engine for AI-assisted development
TidyFactor Skill Architect — Governance Layer & Methodology Engine for AI Coding Agent Skills
Python client for the SPT Models GPU inference platform
AI-powered code review with secret detection, approval workflows, and multi-LLM support
fez's standing agent runtime (buzz-acp's role): persona-backed channel agents with harness invocation, retries, circuit breaker, observer stream, steer/queue, budgets
DeepSeek Harness 插件全家桶聚合安装器:一条命令装齐 壳/会话新窗口/深链/右侧工作台/逐轮产出物卡片/会话撤回(并为引擎补齐撤回所需 truncate 补丁;无功能代码,各功能为独立插件)
OpenCode TUI plugin that renders the session's context-window usage as a colored, segmented bar (cached / prompt / thinking / output / reserved output) in the sidebar
Metrik developer SDK: verify delivery and open, monitor, claim, or reclaim service payment streams.
MemorySync memory for the Vercel AI SDK: recall + persistence middleware for generateText/streamText, standalone helpers, and agent memory tools. Supports AI SDK v6 and v7.
Durable cron automations for DeepSeek Harness
Provider-neutral coding-agent Host and local Node transport primitives.
Portable AI Matrx agent stream protocol, event projection, and safe result-presentation primitives.
Route each Pi session's first task to an appropriate model
Give text-only pi models vision — describe images with a vision model you pick via an interactive picker, then hand off the text description to non-vision models. Includes direct clipboard paste (image-paste integration).
Open-source resume strategist skill for students. Tailors resumes + cover letters in any LLM. Evidence-based. MIT.