All the skills you'll need for AI-powered development with pi-coding-agent. 60+ skills covering research, content, project management, debugging, and more. Includes external-scout for implementation plans, dynamic model providers (OpenRouter, NVIDIA NIM), and comprehensive tooling for modern development workflows.
Agent Claude Code
Software architecture specialist for system design, trade-off analysis, and architecture decision records. Use when designing new features, evaluating technical trade-offs, planning system scalability, or creating ADRs.
Agent Claude Code
TypeScript and frontend build error resolver fixing type errors, module resolution, and config failures. Use when tsc --noEmit or npm run build fails, imports can't be resolved, or tsconfig/webpack/Next.js config errors block development — with minimal diffs and no architectural changes.
Agent Claude Code
Senior code reviewer ensuring high standards of quality, security, and performance in pull requests and diffs. Use when reviewing code changes for bugs, security vulnerabilities, readability, test coverage, or performance regressions before merging.
Agent Claude Code
SDK compliance reviewer validating library usage against official documentation. Use when adding or upgrading SDK integrations, reviewing smart-contract interactions, checking for deprecated APIs, or confirming code follows current SDK best practices and type safety.
Agent Claude Code
PostgreSQL specialist for query optimization, schema design, Row Level Security, and performance review. Use when reviewing SQL for slow queries and missing indexes, designing schemas, adding RLS policies, fixing N+1 patterns, or tuning concurrency and connection management.
Agent Claude Code
Documentation and codemap specialist keeping docs in sync with the actual codebase. Use when generating or updating architecture codemaps, refreshing READMEs after feature or API changes, validating doc links and examples, or auditing which files, endpoints, or dependencies the docs mention.
Agent Claude Code
End-to-end test specialist for creating, maintaining, and executing Playwright tests with artifact and flaky-test management. Use when writing E2E tests for critical user journeys, debugging failures from HTML/JUnit reports, hunting flaky tests, or wiring E2E into CI.
Agent Claude Code
Research external open source projects, clone them to /tmp/, analyze code patterns, and provide findings for planning.
Agent Claude Code
Go build error resolver fixing compile errors, go vet warnings, and linter issues. Use when go build fails, vet or staticcheck reports warnings, a module/import cycle appears, or interfaces don't satisfy — with minimal, surgical changes and no refactoring.
Agent Claude Code
Senior Go code reviewer enforcing idiomatic Go, security, error handling, and concurrency best practices. Use when reviewing Go diffs for SQL/command injection, race conditions, ignored errors, non-idiomatic patterns, or performance issues before merging.
Agent Claude Code
Implementation planning specialist creating comprehensive, actionable step-by-step plans. Use when breaking down a feature request, sequencing implementation phases, identifying dependencies and risks before coding, or planning refactors with minimal disruption.
Agent Claude Code
Senior Python code reviewer enforcing PEP 8, Pythonic idioms, type hints, security, and performance. Use when reviewing Python diffs for injection or eval risks, bare excepts, missing type hints, non-idiomatic patterns, concurrency bugs, or before merging any Python code change.
Agent Claude Code
Refactoring and dead code cleanup specialist. Use when identifying unused code, consolidating duplicates, removing unused dependencies, or performing safe codebase cleanup with deletion tracking.
Agent Claude Code
Code review specialist for quality, correctness, and security analysis.
Agent Claude Code
Rust build/quality error resolver fixing compile errors, clippy warnings, fmt issues, and test failures. Use when cargo build, clippy, fmt --check, or tests fail and you need minimal surgical fixes without refactoring the codebase.
Agent Claude Code
Rust code review specialist for ownership, memory safety, error handling, and idiomatic patterns. Use when reviewing Rust code changes, checking for ownership/borrowing issues, auditing unsafe blocks, or enforcing Rust best practices.
Agent Claude Code
Fast codebase recon that returns compressed context for handoff to other agents.
Agent Claude Code
Security review specialist for web application vulnerability detection and remediation. Use when reviewing code for OWASP Top 10 issues, detecting hardcoded secrets, validating authentication/authorization, or auditing dependencies.
Agent Claude Code
Test-driven development specialist enforcing write-tests-first workflows with comprehensive coverage. Use when developing new features, fixing bugs with regression tests, guiding developers through the red-green-refactor cycle, or ensuring 80%+ unit, integration, and E2E coverage.
Agent Claude Code
General-purpose subagent with full coding capabilities in an isolated context.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: