elieandraos

lab-it

Investigates and validates how an existing system or capability actually works, from real implementation, tests, and current evidence — never conventions or guesses — producing a verified answer, an architecture decision reached with the user, or an approved `plan.md` handed to `plan-it`. Trigger to investigate or explain how a system works, resolve architecture or design decisions for a proposed feature, or synthesize approved findings into `plan.md`. Not for explaining one function, debugging, reviewing a diff, writing API reference docs, or creating, updating, or reviewing an architecture guide — route guide work to `document-it`.

plan-it

Plans a feature end-to-end into GitHub milestone/label metadata and a reviewed, drafted set of issues — classify, scope, draft, review, and create only after approval. Trigger when asked to plan, scope, or scaffold a feature, or to turn an approved plan.md initiative into milestones/issues. When `lab-it` has already produced an approved, initiative-specific `plan.md` section, treat it as canonical input instead of re-deriving decisions from conversation. Also trigger when an unexpected finding needs investigation before deciding whether it's issue-worthy. Only plans, drafts, and reviews GitHub milestones/issues; does not implement code, and does not produce or amend `plan.md` itself (that's `lab-it`'s job).

AI & Automation Listed

ship-it

Delivery-stage skill in the Agentic Engineering pipeline. Checks milestone PR readiness once a milestone genuinely has zero open issues remaining and prepares/creates the milestone PR through authorized human approval; investigates and explains delivery/CI failures on an already-open milestone PR, handing any human-authorized correction to `implement-it`; and, once the human confirms the PR merged and authorizes the post-merge progression, closes the milestone and prepares, publishes, and validates the release. Each entry point checks GitHub's actual current state rather than requiring proof of a prior `implement-it` session. Use when checking whether a milestone is ready for a PR, creating that PR, investigating a delivery/CI failure, checking whether a milestone is ready to close, or releasing a version. Does not implement code, decide what work should exist, or approve/merge the PR — the human retains both.

Data & Documents Listed

document-it

Creates, updates, and reviews explanatory architecture guides in Markdown, a Claude Artifact, or both — reusing sufficient, current, verified understanding already available, and routing to lab-it's investigation method only when evidence is missing or stale. Trigger to document existing architecture, update an existing guide, or review a guide's completeness. Not for debugging or reviewing a diff. Not for investigating a system from scratch, resolving architecture decisions for a proposed feature, or synthesizing plan.md — those stay with lab-it.

review-it

Standalone implementation-review skill in the Agentic Engineering pipeline. Reviews a worktree, branch, or PR against a substantive, evidence-backed checklist and reports verified findings, a scoped clean result, or explicit limitations — never fixing what it finds. Callable standalone against any worktree, branch, or PR with no prior implement-it session or plan-it-authored issue required; also callable by implement-it before Gate 1 and during an authorized delivery correction. Trigger to review implemented code, review a branch or PR, or verify a correction before it's relied on. Reports only — it never fixes application code, applies formatting fixes, commits, pushes, approves a gate, merges, or mutates GitHub or other live/production state; every correction returns to implement-it. Does not own guide review, issue/plan-synthesis review, investigation discipline, or commit-plan review — those stay with document-it, plan-it, lab-it, and implement-it.

Testing & QA Listed

laravel-inertia-stack

Companion conventions for a Laravel + InertiaJS + Vue 3 + Pest stack, additive to Laravel Boost's laravel-best-practices and testing-best-practices skills — never a replacement for them. Activate for Laravel, Inertia, or Pest work on this stack — always alongside the matching Boost skill(s), never alone. See the routing table inside for the specific capabilities covered.

implement-it

Implementation-stage skill in the Agentic Engineering pipeline. Takes any approved GitHub issue satisfying its entry contract — whether `plan-it` drafted it or it already existed some other way — and carries it through working-branch readiness, the implementation itself, verification, semantic commits, issue closure, and dependency-ready recalculation for the next issue. Use when implementing, committing, verifying, or closing an approved issue, checking what's next in a milestone, or performing a human-authorized delivery correction handed back from `ship-it`. Performs the approved implementation itself, consulting the applicable stack companion for implementation knowledge and conventions — it does not own framework-specific conventions, decide what work should exist, or handle milestone PR readiness, PR creation, or release.

AI & Automation Listed

steward-it

Retrospectively investigates an engineering session when work was unexpectedly slow, difficult, repeatedly off course, or exposed a recurring agent-workflow problem. Reconstruct the session from available conversation, tool, repository, skill-trace, context-cost, and outcome evidence; distinguish observed facts from inference; classify the likely cause across methodology, skill, project, stack, prompt, execution, or external limitation; detect repeated patterns; and recommend deliberate improvements without changing canonical guidance automatically. Invoke explicitly for requests such as 'steward this session', 'what happened here?', 'why did this take so long?', or 'is this a recurring problem?'. Not part of the normal lifecycle and not an automatic reviewer of ordinary work.