ai-settings

Use when a request leaves material product, architecture, behavior, or scope decisions unresolved. Do NOT use for approved, well-bounded implementation work.

Use when 2 or more substantial tasks are independent, have disjoint state, and parallel execution will save meaningful time. Do NOT use for coupled or sequential work.

Use when implementing an approved written plan with dependent tasks in the current session. Do NOT use when the plan is missing, unapproved, or naturally parallel.

Use when verified implementation is ready and the user must choose whether to keep, merge, or publish the branch. Do NOT use before applicable checks pass.

Use when receiving code review feedback that may be unclear or technically questionable and needs verification. Do NOT use for an unrequested general code review.

Use when a completed change is high-risk, crosses subsystem boundaries, or the user explicitly requests independent review. Do NOT use by default for low- or medium-risk changes.

Use when the user requests delegated implementation or multiple substantial tasks are genuinely independent without shared state. Do NOT use for coupled or ordinary single-task work.

Use when diagnosing a bug, failing check, unexpected behavior, or production incident before changing code. Do NOT use for known mechanical edits without a failure.

Use when implementing a high-risk behavior change, reproduced regression, or costly contract. Do NOT use by default for docs, styling, config, or low-risk wiring.

Use when feature work needs isolation from the current workspace or before executing a risky implementation plan. Do NOT use when the current workspace is already safe and isolated.

Use when starting a conversation or deciding whether a Superpowers workflow matches the current task. Do NOT use it to auto-invoke skills whose stated triggers do not match.

Use when about to claim work is complete, fixed, passing, or ready to integrate. Do NOT use to repeat checks without a completion claim or concrete doubt.

Use when an approved change spans multiple dependent outcomes and needs a durable implementation sequence before coding. Do NOT use for small, obvious, reversible edits.

Use when creating, editing, or validating reusable agent skills and their trigger behavior. Do NOT use for ordinary project documentation or one-off prompts.

Use when starting a new project from scratch. Scaffolds project structure from public template repo, creates GitHub repo, copies AI settings snapshot, and generates a ready-to-execute deployment runbook for the Yandex Cloud VM. Trigger: user calls /popovs:boilerplate or says "создать новый проект / развернуть проект". SKIP: if the project is already initialized (git repo exists, files present).

Use when the user asks "обнови changelog / добавь запись в changelog", or after a user-visible change is committed. Also trigger automatically after a `feat:` or `fix:` commit. SKIP: for `chore:`, `docs:`, `test:`, `refactor:`, `style:`, `ci:`, `build:` commits unless the change is user-facing.

Use when the user asks "напиши коммит / сгенерь commit message / опиши этот diff as commit". Also trigger automatically before `git commit` if no message was provided. SKIP: if the user already wrote a commit message, or for trivial merge/revert commits.

Use when the user asks "напиши PR / сгенерь PR description / опиши что я делал в этой ветке". Also trigger automatically before `gh pr create` when no `--body` argument was provided. SKIP: if the PR already has a description, or for draft PRs explicitly marked "wip".

Use when the user asks "напиши пост в тг / оформи это для телеграма / сделай tg-пост про X". Target: personal channel or group Telegram posts in Russian, in the channel's established voice. SKIP: posts for other platforms (Twitter, LinkedIn, Habr — use separate skills), formal press releases.