Configures Husky, lint-staged, commitlint, and Python pre-commit

Coordinates project structure: SCAFFOLD new or RESTRUCTURE existing to Clean Architecture

Coordinates logging, error handling, CORS, health checks, and API docs

Calculates Usefulness Score = Impact (1-5) × Probability (1-5) for each test. Returns KEEP/REVIEW/REMOVE decisions based on thresholds (≥15 KEEP, 10-14 REVIEW, <10 REMOVE).

Detects tests that validate framework/library behavior (Prisma, Express, bcrypt, JWT, axios, React hooks) instead of OUR code. Returns findings with REMOVE decisions.

Test suite audit coordinator: automated + manual tests. Delegates to 7 workers (Business Logic, E2E, Value, Coverage, Isolation, Manual Quality, Structure). Output: docs/project/test_audit.md.

Worker that runs existing tests to catch regressions. Auto-detects framework, reports pass/fail. No status changes or task creation.

Universal seed data generator: MIGRATE from ORM schemas or GENERATE from entity definitions to any target format

Coordinates linters, pre-commit hooks, and test infrastructure setup

Generate an execution plan from a spec, with milestones, tasks, and test coverage for all acceptance criteria. Use when the user wants to plan a project, break a spec into milestones, generate tasks from a spec, or create an execution plan. Optionally creates a Linear project.

Migrates a monorepo package from `tsup` to `tsdown`. Use when asked to migrate a package to `tsdown` and the package has used `tsup` thus far.

Generate local PR/branch review artifacts for the current branch vs its base - an in-repo canonical spec (if present) or an inferred review `spec.md`, plus `system-design-review.md`, `code-review.md`, and `walkthrough.md` (via `.agents/skills/drive-pr-walkthrough/SKILL.md`). Writes artifacts to disk (next to the in-repo spec when present, otherwise under `wip/`). Use when the user asks for a local PR/branch review, a code review, a system design review, to "review this branch", or to produce written review docs. Do not modify implementation code.

Create a new transient project workspace under projects/<project>/ (folders + optional stub docs), then hand off to drive-create-spec / drive-generate-plan.

Write an intent-first walkthrough (semantic narrative) of a PR/branch or commit range - the overall purpose, the sequence of conceptual steps, the concrete behavior changes, and links to both implementation touchpoints and tests as evidence. Use during branch/PR review when the user asks for a walkthrough, narrative of changes, semantic diff, intent of commits, or “what changed and why” (not a file-by-file diff recap).

Generate and iteratively refine engineering specs from a description, conversation summary, or blank template. Use when the user wants to create a spec, PRD, product requirements document, write up requirements, or spec out a feature. Also use when summarising a conversation into a spec.

Bumps `biome` package versions (e.g. `@biomejs/biome`) using `pnpm`, aligns `biome.jsonc` files with the new version/s across the repository and runs biome-related checks. Use when required to update `biome` to a newer version - explicitly or implicitly (e.g. after running `pnpm up`, `pnpm update`, `pnpm upgrade` without specific package names).

Creates a GitHub PR with a conventional-commit title and a narrative description for prisma-next. Use when the user wants to create a pull request, open a PR, or submit changes for review.

Detects and fixes accessibility issues in React/Fluent UI webviews. Use when reviewing code for screen reader compatibility, fixing ARIA labels, ensuring keyboard navigation, adding live regions for status messages, or managing focus in dialogs.

Record real OpenAI/Anthropic HTTP back-and-forth (requests + responses, including streaming text/event-stream) and print paste-ready Swift fixtures for SwiftAgent unit tests (ReplayHTTPClient) using the AgentRecorder CLI or HTTPReplayRecorder. Use when adding/updating any provider adapter tests (text, streaming, structured outputs, tool calls), when payload formats change, or when debugging agent loop mismatches by inspecting recorded JSON/SSE payloads.

Azure Cosmos DB performance optimization and best practices guidelines for NoSQL, partitioning, queries, and SDK usage. Use when writing, reviewing, or refactoring code that interacts with Azure Cosmos DB, designing data models, optimizing queries, or implementing high-performance database operations.