SECURITY REVIEW
Not yet assessed
Review the original instructions and requested permissions before installing.
No security review is available for this catalog entry yet.
Implement features and fix bugs from defined tasks while maintaining codebase quality.
Implement features, fix bugs, and keep the codebase healthy and maintainable. Works from a defined PBI, issue, or task description.
Review the original instructions and requested permissions before installing.
No security review is available for this catalog entry yet.
How clearly the skill guides your agent, how complete its workflow is, and how you can check the outcome.
No quality assessment is available for this catalog entry yet.
Original instructions from the publisher’s SKILL.md
# Dev — implement features and fix bugs Implement features, fix bugs, and ensure the codebase remains healthy and maintainable. ## Boundaries - **In scope:** application code, tests, and the validation loop that proves a change works. - **Out of scope:** authoring specs (`spec`), creating Linear issues (`lead`), reviewing the changeset (`critic`), committing and pushing (`ship`). ## Inputs Work from a defined Product Backlog Item with acceptance criteria whenever one exists. Before writing code, check Linear for the assigned issue and `specs/` for the spec covering the feature domain. ## Steps 1. **Load the contract.** Read the PBI or task description, the relevant spec, and the code in the feature area. 2. **Implement.** Make the change, staying inside the acceptance criteria. 3. **Validate.** Run the checks below; all must pass before the work is complete. 4. **Record progress.** Update the Linear issue status and add a comment with completion notes. ## Standards - **Type safety:** use TypeScript strictly. - **Component imports:** explicitly import every component used in `.astro` files — Astro has no auto-imports. - **Content collections:** use the modern `loader: glob()` form. - **Styles:** link CSS through the build, never `<link href="/src/...">`. ## Hard constraints - NEVER use `any` types. - NEVER use the legacy `type: 'content'` for collections. - NEVER output code with broken imports. - NEVER link CSS via `<link href="/src/...">`. ## Sub-agent model policy When parallelizing work into sub-agents, use the cheapest model that fits: - **haiku** — file search, grep, reading files, listing issues, gathering context. - **sonnet** — code edits, writing content, running validation commands, updating Linear issues. - **opus** — ask the user for confirmation before spawning one, and explain why the sub-task needs it. Set the `model` parameter on each sub-agent call accordingly. Default to **sonnet** when unsure. ## Validation - `pnpm check` — type validation - `pnpm lint` — Biome checks - `pnpm test:run` — unit tests ## Task input Apply the skill to the supplied PBI number, issue ID, or task description.