skills/visualize-plan/SKILL.md
Generate a "visual companion" for one or more PRD/TRD markdown files. Long specs become a multi-page mini-site with a sticky side menu (folder of self-contained HTML files); short specs become a single-page HTML. Both follow the Liti mobile design system and a fixed page-shell template so visuals stay consistent across the team.
npx skillsauth add litisaude/garage visualize-planInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Turn a markdown plan into a focused review surface — a multi-page mini-site for long specs, a single page for short ones. Both share the same page shell, design tokens, and snippet library so every spec the team produces looks like a sibling of the others.
/visualize-plan <path-to-md> [<path-to-md> ...] — explicit input(s)./visualize-plan — no args: scan docs/specs/**/*.md from the current branch's diff against main and ask the user which to include.Use when: PRD/TRD has architecture, workflows, state machines, deployment topology, or REST APIs that read better as diagrams than ASCII; or a long markdown is hard to scan; or paired spec + impl-plan need joint navigation.
Don't use for: bug-fix tickets, pure SQL migrations, code reviews (use /review code).
| Mode | When | |---|---| | Mini-site (folder) | md > 300 combined lines · OR ≥1 new REST endpoint · OR plan covers ≥3 of {architecture, workflows, data model, deployment, API} | | Single page (one file) | Everything else |
<source-md-basename>-visual-companion/<source-md-basename>-visual-companion.html-visual-companion.html or folders ending -visual-companion/ are auto-regeneratable — never hand-edit.In a mini-site, (1)–(3) live on index.html, (4) on architecture.html, (5) on roadmap.html, (6) on every page.
reference/page-types.md to see each page's contract. index.html and roadmap.html are mandatory; api.html is mandatory if the plan introduces ≥1 new REST endpoint.templates/shell.html for the consistent page frame (CSS, side-menu structure, footer). Every page in a mini-site uses this verbatim.reference/snippets.md for the canonical building blocks (KPI tile, callout, API endpoint card, SVG arrow markers, lock badge, side menu) — copy verbatim, do not reinvent.reference/design-system.md for the design tokens, typography, SVG conventions, and color semantics.examples/commerce-domain/ when you need a concrete pattern to copy. Treat its CSS / SVG / snippet patterns as the canon — when in doubt, match it.<section> open vs close tags; if mini-site, check every cross-link in the side menu resolves to a sibling file.templates/shell.html — the consistent page frame with all CSS, side-menu skeleton, footer. Slot comments mark substitution points.reference/page-types.md — per-page contracts (what index, architecture, flows, data-model, api, deployment, roadmap must contain + which sections are optional).reference/snippets.md — copy-verbatim HTML/SVG patterns: KPI tile, callout, API endpoint card, method pills, status pills, SVG arrow markers, lock badge, side menu structure.reference/design-system.md — Liti mobile design tokens (:root CSS), typography rules, SVG drawing conventions, color semantics, responsive breakpoints.examples/commerce-domain/ — canonical mini-site for ENG-8165 (~1340 combined md lines across spec + impl plan). Demonstrates all 7 page types. The visual canon — open with Read when unsure.The generated output is meant to be re-generated from the plan, not hand-edited. Re-running on the same input overwrites in place. Persistent custom tweaks belong in a sibling file, not inside the auto-regenerated output.
development
# /review — Composite Code Review Run the appropriate reviewer agents based on the mode and stack. ## Usage - `/review plan` — Review a plan/feature description before coding. Produces a requirements checklist. - `/review code` — Review actual code changes. Produces a violation report. - `/review code backend` — Limit code review to backend reviewers only. - `/review code frontend` — Limit code review to frontend reviewers only. ## Mode: `plan` **When**: Before coding starts. The user has a
development
Maintainer-only workflow for handling GitHub Secret Scanning alerts on OpenClaw. Use when Codex needs to triage, redact, clean up, and resolve secret leakage found in issue comments, issue bodies, PR comments, or other GitHub content.
development
Maintainer workflow for OpenClaw releases, prereleases, changelog release notes, and publish validation. Use when Codex needs to prepare or verify stable or beta release steps, align version naming, assemble release notes, check release auth requirements, or validate publish-time commands and artifacts.
development
Run, watch, debug, and extend OpenClaw QA testing with qa-lab and qa-channel. Use when Codex needs to execute the repo-backed QA suite, inspect live QA artifacts, debug failing scenarios, add new QA scenarios, or explain the OpenClaw QA workflow. Prefer the live OpenAI lane with regular openai/gpt-5.4 in fast mode; do not use gpt-5.4-pro or gpt-5.4-mini unless the user explicitly overrides that policy.