agent-creator
Creates or updates primary and sub-agents in the opencode agentic system. Use when user asks to "create an agent", "add an agent", "update an agent", "modify an agent", "configure an agent", "add a sub-agent", "add a primary agent", or needs to define a new agent role, permissions, or skill access.
commit
Analyzes repository state, proposes commit messages following Conventional Commits, and applies commits after user approval. Use when asked to "commit", "commit changes", "save my work", "create a commit", or "stage and commit".
conversation-memory
Records and retrieves durable user feedback, working preferences, and project-scoped conversation notes in SQLite for future reference. Use when a conversation starts or ends, when user says preferences such as "remember", "don't do that", "next time", "always", or "never", and before applying future work in a directory.
grill-me
Conducts thorough interviews to deeply understand user needs, requirements, and context before any implementation begins. Use when requirements are unclear, assumptions need validation, edge cases need exploration, when the user says "grill me", "ask me questions", "help me think through this", or when the underlying problem isn't fully understood.
legal-advice
Analyzes legal documents and answers legal questions using structured reasoning. Use when the user asks to "review this contract", "check this NDA", "analyze legal risks", "explain this legal clause", "what are my legal rights", or asks for "legal advice" on documents, agreements, or procedures.
playwright-cli
Interactive browser automation using playwright-cli bash commands. Use when user asks to "open a page", "take a screenshot", "check the layout", "verify the UI", "inspect element", "fill out a form", "click a button", "mock API responses", "record a trace", or "check console logs". Also use for debugging design issues, verifying responsive layouts, testing form interactions, inspecting cookies or localStorage, checking network requests, or tracing page performance. Do NOT use for writing Playwright test files (.spec.ts or .test.ts) or running test suites (npx playwright test).
pull-request
Analyzes branch diffs, drafts PR titles and bodies following Conventional Commits, and creates or updates pull requests via GitHub CLI. Use when user asks to "create a PR", "open a pull request", "update PR description", "draft a PR", or "submit changes for review".
review-validation
Validates PR review comments against actual code by analyzing reviewer claims to determine if they are valid or invalid. Use when user asks to "validate review comments", "check if review feedback is correct", "verify PR review", or "are these review comments valid".
review
Performs code review analysis across Quality, Regression, Documentation, and Performance focus areas with severity-classified findings. Use when user asks to "review code", "review this PR", "check code quality", "review changes", or "do a code review".
skill-creator
Creates new Agent Skills (SKILL.md files) following the agentskills.io open standard. Use when user asks to "create a skill", "write a skill", "make a new skill", "build a skill", "scaffold a skill", or needs help with skill structure, frontmatter, progressive disclosure, or skill authoring best practices.