OpenCode skills
How to use Skills in OpenCode: where OpenCode reads them, how skilld installs them, and a curated list with authors. The same Skill runs in Claude Code, Codex, Cursor, and every other Agent skilld targets.
Agents
npx skilld run owner/repo/skillOpenCode is an open-source coding Agent for the terminal. It reads Agent Skills from .opencode/skills in a project and from ~/.config/opencode/skills for every project. skilld gives you a curated set of Skills, each written by a person in their own repository, and one command that puts a Skill in front of OpenCode.
How Skills work in OpenCode
- Project:
.opencode/skills/<skill>/SKILL.md - Global:
~/.config/opencode/skills/<skill>/SKILL.md. IfXDG_CONFIG_HOMEis set, OpenCode and skilld both use$XDG_CONFIG_HOME/opencode/skills.
OpenCode discovers Skills at startup and loads the full file when a task matches.
Install a Skill for OpenCode
npx skilld install owner/repo/skill --agent opencodeThe Skill lands in .opencode/skills. Add --global to install to ~/.config/opencode/skills instead. skilld detects the Agents in your project and on your machine. If it finds only one, omit --agent.
Start a new OpenCode session after an install.
Run a Skill without installing it
npx skilld run owner/repo/skillskilld run prints the Skill and writes no file. Give the command to OpenCode. It reads the Skill and follows it for this session. Nothing lands in your repository, and no lockfile changes.
Curated Skills for OpenCode
OpenCode readers work in existing codebases, so these Skills cover review and refactoring. Skills here come from the curated registry; a person admitted each one.
Resolve PR review bot findings on current PR. Fetches unanswered bot comments, evaluates each finding, fixes real bugs, dismisses false positives, replies to every comment, and watches for new findings until bots go quiet.
Use when completing tasks, implementing major features, or before merging to verify work meets requirements
Use Sentry MCP to discover, triage, and fix production issues with root-cause analysis. Use when asked to fix Sentry issues, triage production errors, investigate error spikes, or clean up Sentry noise. Requires Sentry MCP server. Triggers on "fix sentry", "triage errors", "production bugs", "sentry issues".
Use when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementation
Review a git diff or explicit file scope for reuse, code quality, efficiency, clarity, and standards issues, then optionally apply safe Codex-driven fixes. Use when the user asks to "simplify code", "review changed code", "check for code reuse", "review code quality", "review efficiency", "simplify changes", "clean up code", "refactor changes", or "run simplify".
Surgical code refactoring to improve maintainability without changing behavior. Covers extracting functions, renaming variables, breaking down god functions, improving type safety, eliminating code smells, and applying design patterns. Less drastic than repo-rebuilder; use for gradual improvements.
Use when writing a PR body.
Keeps findings that affect correctness, security, data, API compatibility, or performance. Drops speculative issues and style suggestions before publishing.
All 69 Skills in Review and refactoring
OpenCode can also search the registry itself, with the skilld Skill or the skilld MCP server. See the setup
Read before you run. A Skill is instructions for your Agent. Open the source file first. What skilld verifies, and what it never claims .