agent-primer

Use when starting any conversation or task so the agent checks for relevant skills before acting, including before clarifying questions, exploration, implementation, review, or documentation work.

Use when reviewing a diff, pull request, patch, or proposed implementation for bugs, regressions, missing tests, security issues, performance problems, or maintainability risks.

Use when the user wants a new capability, asks whether a skill exists, asks to find or install skills, or describes a recurring workflow that may already have an installable skill.

Use when choosing, adding, configuring, or reviewing MCP servers, MCP client config, tool permissions, remote server URLs, local stdio commands, or MCP-related security tradeoffs.

Use when code or config touches authentication, authorization, user input, secrets, tokens, database access, file access, external APIs, MCP servers, browser automation, dependency installation, or deployment.

Use when creating, updating, pruning, or reviewing a SKILL.md so recurring workflows become concise, reusable agent skills.

Use when implementing, fixing, refactoring, or reviewing code and you need a disciplined verification loop with builds, tests, type checks, linting, formatting, or manual checks.