gotools

Clone and access external git repositories the agent needs to read, reference, or work with. Use this skill whenever you need code, configs, or documentation from another repository — whether to inspect an implementation, check a dependency, read a shared library, cross-reference API contracts, or pull in any external codebase. Also use when a task mentions a repo you don't have locally, when you need to look something up in another project, or when the user asks you to check, clone, or fetch a repository.

Analyze a codebase for bugs, prioritize by severity, then hand off the most critical one to $patch for fixing. Use when asked to find bugs, hunt for issues, or proactively improve code correctness.

Proactively scan a codebase for missing or outdated documentation, prioritize by user impact, then hand off the most critical gap to $patch for improvement. Use when asked to find documentation gaps, audit docs freshness, or improve developer experience.

Proactively scan a codebase for reliability issues, prioritize by failure impact, then hand off the most critical one to $patch for hardening. Use when asked to find reliability problems, audit error handling, or improve failure resilience.

Proactively scan a codebase for security vulnerabilities, prioritize by exploitability and impact, then hand off the most critical one to $patch for remediation. Use when asked to find security issues, audit for vulnerabilities, or harden a codebase.

Proactively scan a codebase for code smells and structural complexity, prioritize by impact on maintainability, then hand off the most critical one to $patch for simplification. Use when asked to find code smells, reduce complexity, or improve code structure.

Proactively scan a codebase for style inconsistencies and convention violations beyond what linters catch, prioritize by confusion risk, then hand off the most impactful one to $patch for cleanup. Use when asked to find style issues, audit naming conventions, or improve codebase consistency.

Proactively scan a codebase for test coverage gaps, prioritize by risk of undetected regression, then hand off the most critical gap to $patch for coverage. Use when asked to find untested code, improve coverage, or audit test quality.

Execute a small, scoped task end-to-end — branch, implement, test, verify, commit, and open a PR. Use when you have a well-defined task (bug fix, docs improvement, small feature, refactor) and need to land it as a complete unit of work.

Create, maintain, and execute ExecPlans for complex or multi-step work that spans many files or repositories. Use when work needs a self-contained living plan with milestones, decision tracking, concrete commands, validation criteria, and restartable context.

Perform structured code review focused on correctness, test coverage, and regressions, with findings prioritized by severity and grounded in file/line evidence. Use when asked to review, audit, assess risk, or evaluate change quality before merge or commit.