OrchestrUI
AGENTS.md1 starsChanged 3 months agoMITView on GitHub
Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.
Of the AGENTS.md files we read, the 31,542 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repository7 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- ui-library-routerSkill
- ui-orchestratorSkill
- ui-quality-auditSkill
- ui-library-routerSkill
- ui-orchestratorSkill
- ui-quality-auditSkill
The file
We split this AGENTS.md into the rules it gives an agent
11 rules. About 2 are common, 5 are distinctive, and 4 are very specific to this project.
≈2
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 8
Verify package names, registry URLs, component IDs, MCP setup and licenses before changing integrations.
- Very specific: line 12
Never vendor or redistribute the React Bits component collection from OrchestrUI.
- Common: line 13
Never copy paid/Pro content, credentials, authenticated registry material or license keys.
- Very specific: line 14
Treat Rive runtime licensing separately from .riv asset rights.
- Distinctive: line 15
Do not imply upstream endorsement or partnership.
- Distinctive: line 20
Use the fewest libraries needed.
- Very specific: line 23
Add Anime.js only when built-in/CSS motion is insufficient.
- Distinctive: line 24
Use Rive only for meaningful interactive vector/state-machine graphics with known asset rights.
- Distinctive: line 25
Avoid accidental daisyUI/shadcn base-system conflicts.
- Very specific: line 36
Read .agents/skills/ui-quality-audit/SKILL.md before completion.
- Common: line 43
no secrets, paid/Pro source or mirrored component collections
≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md 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.
Files
OrchestrUI repository instructions
Mission
OrchestrUI is an agent-facing UI/UX orchestration layer for Kokonut UI, React Bits, daisyUI, Bklit UI, Anime.js, Rive, and Magic UI. It is not a vendor dump.
Source of truth
- Current official upstream docs/repos are authoritative.
- Verify package names, registry URLs, component IDs, MCP setup and licenses before changing integrations.
- Keep
catalog/*.jsonmachine-readable.
Legal guardrails
- Never vendor or redistribute the React Bits component collection from OrchestrUI.
- Never copy paid/Pro content, credentials, authenticated registry material or license keys.
- Treat Rive runtime licensing separately from
.rivasset rights. - Do not imply upstream endorsement or partnership.
Routing principles
- Inspect the host project first.
- Existing design systems win unless migration is explicit.
- Use the fewest libraries needed.
- Give each major UI role one owner.
- Bklit is the preferred chart/data-viz specialist.
- Add Anime.js only when built-in/CSS motion is insufficient.
- Use Rive only for meaningful interactive vector/state-machine graphics with known asset rights.
- Avoid accidental daisyUI/shadcn base-system conflicts.
- Accessibility, responsiveness, performance and coherence outrank novelty.
Workflow
- Read
.agents/skills/ui-library-router/SKILL.md. - Inspect the host stack and produce a short selection plan.
- Read
.agents/skills/ui-orchestrator/SKILL.md. - Retrieve current official upstream information only for selected libraries.
- Harmonize typography, spacing, radius, color, surfaces, shadows and motion.
- Implement responsive and accessible behavior.
- Run host-project checks.
- Read
.agents/skills/ui-quality-audit/SKILL.mdbefore completion.
Repository gates
npm run validatenpm test- exactly seven core libraries
- valid skill frontmatter
- no secrets, paid/Pro source or mirrored component collections
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
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.

