danielvm-git/bigpowers

Align Grid builds Müller-Brockmann modular grid webpages with CSS-variable-driven columns, baseline rhythm, and optical alignment. Use it when you need a visible, load-bearing grid that can be toggled and verified with a Puppeteer harness.

Assess Impact analyzes the blast radius of a proposed change before code is written. It maps dependents, affected stories, and test coverage, then writes specs/IMPACT_LATEST.md.

Audit Code runs a self-review checklist before requesting review or committing. It checks security, tests, types, conventions, and code quality, then produces pass/fail results.

Audit Plan evaluates an incoming project plan against bigpowers principles and conventions, then records gaps and a READY or NOT READY verdict before build work starts. Use it when a new project arrives or before seeding conventions in an unfamiliar codebase.

Build Epic orchestrates a nine-step build cycle for one epic, from threat modeling and planning through TDD, verification, audit, commit, and release. Use it to resume or run release work with state tracked in YAML files.

Change Request adds or reorders release requirements against specs/release-plan.yaml and epic capsules. Use it when a new requirement arrives mid-release or the plan needs reprioritization.

Commit Message reviews working-tree changes and drafts a Conventional Commits title, body, and release bump. It is used when preparing a commit message or checking semantic-release and semver impact before git commit.

Compose Workflow chains multiple BigPowers skills into a documented workflow recipe saved in specs/. Use it when a project needs a repeatable non-standard sequence or a command-mappable playbook.

Context7 MCP fetches current library documentation through Context7 instead of relying on training data. Use it for framework, API, setup, and example questions about libraries like React, Next.js, or Prisma.

Craft Skill helps create new Bigpowers skills with the required structure, triggers, and bundled resources. Use when you need to write, build, or package a new skill for the Bigpowers lifecycle.

Craft Skill helps create new bigpowers skills with proper structure, progressive disclosure, and bundled resources. Use when you need to write or build a new skill for the bigpowers lifecycle.

Deepen Architecture finds shallow modules and refactoring opportunities in a codebase using architecture docs and ADRs. Use it when improving testability, reducing coupling, or making the codebase easier for agents to navigate.

Define Language extracts a DDD-style ubiquitous language glossary from a conversation and saves it to specs/UBIQUITOUS_LANGUAGE_LATEST.md. Use it when you need to standardize domain terms, flag ambiguities, or build a glossary.

Define Success is a tombstoned stub that redirects users to plan-work after a rename or merge. Use it only during the transition period before the stub is removed.

Delegate Task runs one subagent on a complex task and reviews its work in two stages before accepting it. Use it when a task needs careful oversight, clear verification, and no parallel execution.

Deploy orchestrates a build-to-production pipeline: build the artifact, verify it exists, deploy, wait for completion, and run a baseline smoke test. Use it after tests on main or in CI/CD to push a release safely.

Design Interface generates multiple radically different module interfaces in parallel and compares their trade-offs before choosing one. Use it when designing an API, exploring module shapes, or following a “design it twice” approach.

Develop TDD guides test-driven development with a red-green-refactor loop using vertical slices. Use it for features or bugs when you need to plan first, write tests before code, and verify behavior through public interfaces.

Diagnose Root runs a four-phase root cause analysis for confirmed bugs: reproduce, isolate, hypothesize, and verify. Use it when the cause is unclear and you need evidence before proposing a fix.

Diagnose Stall identifies why agent orchestration stopped making progress in loops, dispatches, or plan execution. Use it when output goes quiet, a subagent never returns, or a checkpoint is overdue.

Dispatch Agents runs multiple subagents in parallel on independent tasks. Use it when work can be split into decoupled pieces and speed matters, with explicit synchronization points and no inter-task review gate.

Edit Document restructures existing documents for clarity, flow, and tighter prose while preserving intent and accuracy. Use it for specs, READMEs, articles, or technical writing that already exists and needs revision.

Evolve Skill benchmarks a skill, identifies failing scenarios, proposes a minimal change, edits the skill, reruns the benchmark, and records an ADR. Use when a skill underperforms or a stocktake reveals a systemic gap.