AgentSystemLabs

Show Mermaid diagrams in Mission Control's diagram viewer instead of dumping raw syntax in the terminal. Use when the user asks for a diagram, flowchart, sequence diagram, architecture sketch, state machine, ER diagram, or any visual that Mermaid can render — and whenever a diagram would clarify a complex flow. Requires a Mission Control agent session (MC_API_URL and MC_API_TOKEN are injected automatically). Not for ASCII art in chat, image generation, or projects running outside Mission Control.

Save durable project knowledge to Mission Control's Recall (project memory) so future sessions start already knowing it. Use when you discover or decide something worth remembering about THIS project — an architecture fact, a decision and its rationale, a convention, a stack detail, a glossary term, a known issue, or a useful discovery ("X lives in Y", "Z is generated"). Requires a Mission Control agent session (MC_API_URL, MC_API_TOKEN, MC_TASK_ID are injected automatically). Not for transient to-dos, run-specific notes, or projects running outside Mission Control.

Add a Playwright end-to-end test for a user-facing flow — sign-in, form submit, multi-step wizard, payment, navigation, etc. Detects existing Playwright setup and inherits its config (baseURL, projects, fixtures, auth state); installs Playwright via the official `npm init playwright@latest` flow only with explicit user approval. Wires one smoke test that hits the real running dev server before generating the full flow. Prefers role-based selectors (`getByRole`, `getByLabel`) over CSS/test-ids; uses `expect`-with-auto-retry assertions; adds an auth-state fixture for tests that need a logged-in user. Trigger phrases — "add an e2e test", "playwright test", "browser test for X", "test the signup flow", "/add-e2e-test", "end-to-end test". Skip for — pure unit-testable logic (use write-tests), API-only flows (use write-tests with supertest/fetch), and projects where the user explicitly does not want a browser harness.

Sibling to polish-ui's loading-state checklist. After data fetching is wired in a route or component, verify and auto-fix the two non-loading states most often forgotten — empty (the request succeeded but returned zero items / null) and error (the request failed). Empty states get a clear message + primary call-to-action. Error states get a human-readable message + retry affordance + (in dev) a way to surface the actual error. Uses the project's existing UI primitives — shadcn `<EmptyState />` if present, project conventions otherwise. Does NOT cover loading skeletons. Trigger phrases — "add empty state", "what if this list is empty", "add error state", "handle the error case", "/add-empty-error-states", "error UI", "no results UI", any time a route was just wired with `useQuery` / `useSuspenseQuery` / loader. Skip for — pure static pages, mutations (useMutation has different patterns), modals/dialogs whose empty state is the parent's responsibility.

End-to-end workflow for planning and shipping a new feature in an existing codebase. Phases — clarify → explore → design → mandatory plan-approval gate → implement (subagent fan-out where useful) → verify → gated reviews (code + duplication always; security/performance/contracts/concurrency/observability/data integrity when gates trigger) → automated tests. Also enforces UI-convention parity: any new instance of a recurring UI surface (modal, dialog, drawer, form) must match sibling conventions for hotkeys, kbd hints, focus, and loading states. Accepts `mode=fast|balanced|production` to control depth (default: production); also accepts `include=<phases>` / `skip=<phases>` overrides. Use when the user asks to add, build, ship, or implement a new feature: "add a feature", "build this", "implement X", "new feature", "plan and build", "ship this". Skip for bug fixes (use fix-bug), pure refactors, enum/state renames (use realign), or one-line tweaks.

Generate and review a Drizzle (or Prisma / Knex / raw SQL) migration safely — classifies the change as additive, mutating, or destructive; for mutating/destructive changes, splits into a multi-phase plan (deploy code that tolerates both shapes → migrate → deploy code that requires the new shape). Catches the landmines: NOT NULL on a populated column without backfill, dropping a column still referenced by old code, adding a unique constraint with existing duplicates, large-table index without `CONCURRENTLY`, type changes that rewrite the table, default values evaluated at migration time vs. row time. Trigger phrases — "add a migration", "schema change", "add column to X", "drop column X", "rename column", "add index", "/add-migration", "drizzle migration", "alter table". Skip for — pure code changes with no schema impact, seed-data-only changes, and infra-level changes (DB version upgrades, replication setup).

After a bug fix lands (or is staged), generate a regression test that fails on the pre-fix code and passes on the post-fix code — pinning the bug so it can't silently return. Walks the agent through reproducing the failure first (against the pre-fix state via `git stash` or temporary revert), writing the assertion that captures it, then verifying the test goes red without the fix and green with it. Trigger phrases — "add a regression test", "pin this bug", "write a test for the fix", "/add-regression-test", "lock in this bug fix", "make sure this doesn't come back", or invoke automatically after a /fix-bug session that produced a code fix. Skip for — fixes whose behavior change is already covered by existing passing tests, pure refactors with no behavior delta, fixes to test code itself, and unreproducible heisenbugs.

Address every unresolved review comment on a GitHub pull request — locate the PR (current branch, PR URL, PR number, or issue number with linked PR), fetch both inline review comments and PR-level issue comments, then for each thread ground the fix in the branch's commits and diff, edit the code, make ONE commit per comment, reply to the thread, and resolve it. Trigger phrases — "address PR comments", "address review comments", "fix review feedback", "respond to PR review", "/address-pr-comments", "address the comments on PR

Generate an image from a text prompt using OpenAI's gpt-image-2 API and save it into the project. Use when the user asks to "generate an image", "make a hero image", "create an og image", "generate a logo/illustration/screenshot placeholder", "I need an image of X", or "/gpt-image-2 <prompt>". After the image is on disk, follow whatever the user asked you to do with it next (wire it into a component, set as og:image, drop into public/, etc.). Skip for: editing existing images (this skill only generates), audits of existing images (use optimize-page-images), or when no API key is available and the user does not want to paste one.

One-shot "ship it" — split a dirty working tree into logically grouped commits, create/push the branch if needed, and open a GitHub pull request with a senior-engineer-quality title, high-level (non-technical) description, auto-applied labels matched against the repo's existing label set, and "might address

Cut a Mission Control desktop release — bump package.json (must match the git tag), commit, create an annotated v-tag, push to trigger release.yml CI. Prefer letting auto-tag-release.yml patch-bump on merges to main; use this skill for major/minor, hotfixes, or when automation was skipped with [skip release]. Read references/mission-control-release.md for the full CI pipeline, academy approval gate, and version alignment rules.

Save durable project knowledge to Mission Control's Recall (project memory) so future sessions start already knowing it, AND navigate this project's indexed code graph. Use when you discover or decide something worth remembering about THIS project — an architecture fact, a decision and its rationale, a convention, a stack detail, a glossary term, a known issue, or a useful discovery ("X lives in Y", "Z is generated"). Also use when you need to LOCATE or READ code — where a symbol is defined (and its verbatim source), what calls it, or what a change would impact — via the graph_search / graph_node / get_neighbors / impact_of / shortest_path MCP tools instead of grepping. Requires a Mission Control agent session (MC_API_URL, MC_API_TOKEN, MC_TASK_ID are injected automatically). Not for transient to-dos, run-specific notes, or projects running outside Mission Control.