karlorz/agent-skills/design-guide-usage

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Record privacy-safe design-guide usage evidence. Use for design-guide usage status or promotion-review readiness.

Install

npx vskill@latest install karlorz/agent-skills/design-guide-usage

SKILL.md

Design Guide Usage

Use the bundled recorder for every operation. Do not recreate its validation, marker, locking, atomic-write, or readiness logic manually.

Resolve The Recorder

Resolve the directory containing this loaded SKILL.md, then use:

text
<skill-directory>/scripts/design-guide-usage.js

Do not derive the path from the current working directory. Do not search arbitrary repository, home-directory, or plugin-cache descendants. If the script is missing, report the expected skill-relative path and stop.

Report Status

Run:

bash
node "<skill-directory>/scripts/design-guide-usage.js" status --json

Let the recorder resolve the active vault through skillwiki path. Pass --vault <path> only when the user deliberately supplies an override.

Summarize the counts, missing signals, eligible, and next_action. Explain that eligibility only opens a human promotion review and performs no copy, package, publication, installation, tag, or release action.

Record Evidence

  1. Confirm the observation came from a meaningful frontend task where

design-guide materially influenced the work. Do not count installation, recorder implementation, tests, or baseline validation.

  1. Gather the explicit evidence fields documented by the recorder's --help.

Use portable project or project-class identities and concise reviewed summaries; do not copy private application source.

  1. Replace sensitive values with [REDACTED:<kind>]. Never persist live

credentials, tokens, passwords, cookies, bearer headers, private keys, customer data, or unrelated private details.

  1. Write the reviewed JSON to a bounded temporary file or send it through

standard input.

  1. Run record --dry-run and show the proposed entry and readiness delta.
  2. Run the write without --dry-run only when the user explicitly authorizes

the write. A request that clearly says to record the reviewed observation is authorization; a request to preview, draft, inspect, or discuss it is not.

  1. Report the evidence identifier, target, updated use count, eligibility, and

next action. Remove any temporary evidence file containing private context.

Example command shape:

bash
node "<skill-directory>/scripts/design-guide-usage.js" record \
--input "<reviewed-json-file>" \
--dry-run

Use the same command without --dry-run only after write authority is clear.

Preserve The Boundary

  • Keep recording operator-invoked and attended.
  • Never scan Codex, Claude, Cursor, lifecycle-hook, shell-history, transcript,

or session stores to infer usage.

  • Never infer usage because either skill is installed.
  • Never append directly to the Markdown log as a fallback after recorder

failure.

  • Never change PATH, install a global executable, or copy into a versioned

plugin cache.

  • Never modify or promote design-guide as a side effect.
  • Fail closed on unresolved vaults, likely secrets, malformed markers,

duplicates, path escapes, lock contention, or concurrent changes.

Security

  • Security scanPassed, 90 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

Repository stars
1
Repository forks
1
7-day trend score
0

Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/karlorz/agent-skills/design-guide-usage/badge)](https://verified-skill.com/skills/karlorz/agent-skills/design-guide-usage)

Something wrong with this skill or its page? Tell us.