GitHub collection
arcasilesgroup/ai-engineering
Imported from GitHub — is this yours?
25 skills imported from this repository.
- Stars
- 65
- Forks
- 3
- License
- Apache-2.0
Skills (25)
ai-audit-code
Finds the simplest working solution for any code change, review, or design; favors YAGNI, stdlib, and native features. Trigger: "minimal", "lazy", "simplest", "do less", or complaints about bloat/over-engineering. Supports lite/full/ultra. Not for prose.
ai-audit-design
Use when a web interface needs its visual defects found and fixed with measurements rather than opinions — misaligned rows, ragged card interiors, touch targets, rendered contrast over gradients and images, text printing over text, Gestalt proximity, divider lines that should be
ai-audit-docs
Audits docs against repo state for dead links, false claims, stale versions, and nonexistent commands. Input: paths or default docs/, README.md, AGENTS.md. Output: ranked findings only; no auto-fix. Trigger: "audit docs", "verify docs", or "documentation review".
ai-brainstorm
Aligns fuzzy ideas before research or planning: interview, classify, and capture the approved intent. Trigger: "handshake", "align on this idea", "my vision", or a new feature/component/behavior. Do not use as a preflight for Direct/light work with a clear outcome. An aligned bui
ai-codegraph
Uses local CodeGraph MCP for symbol relationships, call chains, and blast radius. Trigger: "how does X work", "what calls Y", "what depends on", or "what breaks if changed". Not for text search (rg) or file discovery (glob).
ai-debug
Finds the root cause at file:line and writes a failing check before the fix; resolves merge and rebase conflicts by intent. Trigger: "it's not working", "CI is failing", "why does this fail", or "rebase failed". Not for coverage (/ai-verify), repo tours (/ai-explore), feature des
ai-design-md-planner
Interviews the user about their product and visual taste, commits to a specific referenced aesthetic direction, and writes a DESIGN.md design system file in Google's DESIGN.md format — then audits it with the official `@google/design.md` CLI until it lints clean. Use this skill w
ai-explore
Answers repo questions with file:line evidence and tours unfamiliar modules. Trigger: "where does X live", "how does this work", "walk me through", or "onboard me". Not for external evidence (/ai-research), failures (/ai-debug), or diff correctness (/ai-verify).
ai-issue-report
Reports reproducible faults as a local, allow-listed draft scanned for machine paths, PII, and secrets; shows exact bytes and SHA-256 before confirmation. Trigger: "report this bug", "file an issue", "send this upstream", or "write the incident report". Diagnose own-code failures
ai-note
Saves and searches hard-won, non-obvious findings as committed Markdown, stamped with the commit and affected file patterns for stale detection. Trigger: "save this", "remember this", or "what did we find about". Not for decisions (/ai-brainstorm) or user docs (/ai-write).
ai-orchestrator
Build a whole feature end-to-end through subagents. It plans gated checkpoints, writes tests and prototypes, then loops each checkpoint through implement → behavior gate (CLI/code tests, UI e2e last) → UI gate (live vs prototype) → adversarial review gate until every checkpoint p
ai-pr
Open or finish a pull request from the current branch in the current checkout or selected worktree. Detect the target branch without assuming a forge or branch name, keep the published diff explicit, watch checks and review, and merge only when the request says to finish, land, o
ai-prototype
Build a single-file, clickable HTML prototype of a UI screen or flow, following the design directions given in the request. Use when the user asks to prototype, mock up, or sketch a UI (e.g. "/ai-prototype order list, dense, neutral palette", "mock up the settings page").
ai-read-docs
Checks primary official docs for versioned/external behavior: packages, SDKs, APIs, CLIs, cloud services, current behavior, API drift, setup/config, or high-stakes auth, security, billing, data, migration, deployment, compliance, or privacy. Not general external research (/ai-res
ai-refresh-project
Brings a governed repository's own guidance back in line with its tree: AGENTS.md and the PRD, the docs/ folder, stale ai-eng artifacts, orphan branches and worktrees, and the governance files (layer rules, config, overrides). Manual only. Reads the tree, proposes, and stops; the
ai-research
Finds external evidence with numbered citations; marks unsupported claims unsourced and names unavailable tools. Trigger: "state of the art", "compare options", "find sources", "is this still true", or "what do the docs say". Not for repo answers (/ai-explore), bugs (/ai-debug),
ai-review-ui
Compare the live UI of the web app against its HTML prototype in .ai-engineering/workflow/prototypes/ and report the visual and behavioral gaps. Screenshots both, then two parallel reviewers (subagents, or `claude -p`) diff them, one for visuals and one for behavior. Use when the
ai-security
Audits code for exploitable vulnerabilities with independent validation and schema-checked findings. Trigger: milestone security audits, "security review", "find security bugs", or "pen-test". Reports real impact, not theory. Not for one change's security review (the security-aud
ai-stress-test
Test system behavior under extreme load to find breaking points, capacity limits, and failure modes. Covers stress, spike, and breakpoint testing with k6 (primary), JMeter, or Locust. Refuses unbounded concurrency and production saturation without explicit approval. Trigger for "
ai-test-planner
Plan and write the tests for a feature, bottom-up. Everything that code or the CLI can verify (pure logic, API endpoints, DB constraints, auth/role rules) is tested without a browser. UI end-to-end tests come last and cover only what the lower layers can't. Use when the user asks
ai-verify
Use when work has to be verified — "verify the feature", "did we build what was asked", "review this diff", "is it ready to commit", or when the user wants over-engineering findings or a debt ledger. Choose the right verification tier and produce verdicts with evidence, not impre
ai-visual-recap
Build an optional self-contained HTML recap when the user manually invokes this skill. The output is a local file on the feature branch.
ai-write
Writes technical documentation for this repository: a README, a wiki page, product documentation, API docs or a technical post. Applies the framework's single writing standard (`references/documentation-writer.md`) and verifies every document against the tree. Trigger for "write
ai-writing-behavior
Authors and revises Agent Behavior BEHAVIOR.md specs that capture recurring, judgeable agent conduct. Use when creating, reviewing, splitting, grouping, or calibrating behavior specs, or when translating trace failures and runtime guidance into durable behaviors. Trigger for "wri