Browse Skills

Project Principles defines mandatory code quality and architectural standards for the vm0 project, including YAGNI, strict TypeScript checking, avoiding defensive programming, and zero lint violations. Use it before and during development to keep the codebase simple and maintainable.

Tech Debt scans a codebase for technical debt patterns such as large files, any usage, lint suppressions, and unsafe test mocks, then creates issues to track the findings.

Testing provides comprehensive testing patterns and anti-patterns for writing and reviewing tests. Use it when writing tests, reviewing test code, or investigating test failures.

Verify Product Image checks a generated static ad image opens, matches the requested dimensions, and shows the correct product or subject. It also flags garbled text, wrong branding, and severe artifacts, then records the result in verification.md.

Chorus Code Reviewer reviews the aggregate code change behind a whole feature after every individual task passes, then posts a single structured VERDICT comment on the Idea. Use it as the final read-only adversarial gateway before an Idea's code ships in the AI-DLC pipeline.

Chorus Proposal Reviewer audits PRD and task drafts against the originating Idea and posts a single structured VERDICT comment. Use it when reviewing Chorus proposals for completeness, specificity, dependencies, and integration checkpoints.

Chorus Task Reviewer fetches a Chorus task, its acceptance criteria, and proposal documents, then independently verifies the implementation and posts a structured VERDICT comment. Use it to review submitted work in an AI task management harness.

Code Reviewer performs the final ship-time review of a whole Chorus Idea, checking cross-task integration, architecture, security, regressions, and feature-level test coverage after all individual tasks pass. Use it after the last task of an idea is verified to post a VERDICT comment.

E2E Verification drives the running Chorus dev server in a real browser via Playwright MCP to confirm UI changes end-to-end. Use it when an acceptance criterion says the user sees X and only a real browser can confirm it.

Adversarial Spec Review constructs a skeptical reviewer that attacks a spec before any code is written, using evidence from the codebase, research, and live best-practices. Use it before high-blast-radius builds to refute unsound plans and harden verification.

Backprop traces a bug or failing test to its root cause, then appends a §B entry and usually a §V invariant to SPEC.md so recurrence is impossible. Use it after test failures, bug reports, post-mortems, or VIOLATE flags.

Build implements tasks from a SPEC.md using a single-threaded plan-then-execute loop. It tracks task status, runs the verification command, and on failure invokes backprop to update spec invariants before retrying.