agentleFS
Sign inStart free
IgorGanapolskyIgorGanapolsky/ThumbGate/.agents/skills/openui-catalog-compose-honesty/SKILL.md

openui-catalog-compose-honesty

Skill29 starsChanged 7 months agoMITView on GitHub

OpenUI FORMAT steal: catalog-compose-only, root-first streaming, repair-before-claim. Audit agent-composed UI against an allowlisted component catalog; never clone OpenUI, Thesys Gateway, or a generative-UI SKU. Slash: /openui-catalog-compose-honesty.

We split this skill into the rules it gives an agent

9 rules. About 0 are common, 4 are distinctive, and 5 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 5

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Keep agent-composed UI honest: only allowlisted components, root-first streams, repair before any ready/done claim.

    line 12
  2. Very specific:

    Never install @openuidev/cli, OpenUI Gateway, or Thesys Observability as ThumbGate

    line 15
  3. Very specific:

    Never invent components outside the catalog or eval / new Function generated UI

    line 16
  4. Very specific:

    Never claim UI/compose ready without --repair --claim-ready evidence

    line 17
  5. Distinctive:

    Never dual-edit a sibling OpenUI/generative-UI WIP PR

    line 18
  6. Distinctive:

    Check sibling WIP for OpenUI/generative-UI PRs — do not duplicate.

    line 55
  7. Distinctive:

    Write or load a component catalog JSON.

    line 56
  8. Very specific:

    Run doctor with --catalog + --stream (+ --repair).

    line 57
  9. Very specific:

    Before any UI/compose done claim, re-run with --claim-ready --strict.

    line 58

≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

OpenUI Catalog-Compose Honesty

Goal

Keep agent-composed UI honest: only allowlisted components, root-first streams, repair before any ready/done claim. Steal OpenUI FORMAT — do not clone the product.

Constraints

  • Never install @openuidev/cli, OpenUI Gateway, or Thesys Observability as ThumbGate
  • Never invent components outside the catalog or eval / new Function generated UI
  • Never claim UI/compose ready without --repair --claim-ready evidence
  • Never dual-edit a sibling OpenUI/generative-UI WIP PR
  • ECI: no net-new generative-UI SKU

Reference

  • https://www.openui.com/ (source FORMAT)
  • Gates: require-catalog-compose-only, require-repair-before-compose-claim
  • Script: scripts/openui-catalog-compose-honesty.js
  • Docs: docs/agents/openui-catalog-compose-honesty.md

Examples

# Clean catalog stream → ready
npx thumbgate openui-catalog-compose-honesty \
  --catalog=catalog.json \
  --stream=compose.txt \
  --repair --claim-ready --json

# Unknown component / missing root → fail
npx thumbgate openui-catalog-compose-honesty \
  --catalog='{"components":["Stack"]}' \
  --stream=$'header = Ghost()\n' \
  --json

Catalog:

{ "components": ["Stack", "Card", "Table"] }

Stream (root first):

root = Stack([header, body])
header = Card()
body = Table([row])

Procedures

  1. Check sibling WIP for OpenUI/generative-UI PRs — do not duplicate.
  2. Write or load a component catalog JSON.
  3. Run doctor with --catalog + --stream (+ --repair).
  4. Before any UI/compose done claim, re-run with --claim-ready --strict.
  5. Wire gates from config/gate-templates.json Agent Honesty pair.

Rubric

  • ok=true when doctor status is ready with repairClean=true on the claimed stream
  • ok=false on unknown components, missing root, arbitrary-code markers, or OpenUI SKU clone signals
  • evidence: --json report in the same turn

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free