dev-ikdev-ik/ai-design-rules/skills/knowledge-graph-architect/SKILL.md

knowledge-graph-architect

Skill3 starsChanged 37 days agoMITView on GitHub

Maintain the AI Design Context knowledge graph. Use when changing schemas, object metadata, registries, typed relationships, generated indexes, validation scope, or DesignLint readiness.

In this repository21 other files it gives its agents.
- AGENTS.mdAGENTS.md
- starter-kit/AGENTS.mdAGENTS.md
- accessibility-checkSkill
- accessibility-reviewerSkill
- agent-contextSkill
- design-evidence-researcherSkill
- design-reviewerSkill
- design-system-architectSkill
- design-understandSkill
- information-architectSkill
- interaction-designerSkill
- mobile-ux-expertSkill
- motion-designerSkill
- performance-reviewerSkill
- product-designerSkill
- prompt-architectSkill
- reference-driven-designSkill
- responsive-checkSkill
- ux-reviewerSkill
- visual-designerSkill
- visual-qaSkill

The file

We split this skill into the rules it gives an agent

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

≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
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

Distinctive:
Keep objects, metadata, generated indexes, relationships, and validation behavior consistent as the repository grows.
line 14
Distinctive:
Check ID, alias, slug, status, maturity, and relationship conventions.
line 41
Distinctive:
Verify relationship direction and type.
line 42
Distinctive:
Check generated indexes and graph diagnostics.
line 43
Very specific:
Run or request npm run check.
line 44
Distinctive:
Avoid changing content meaning while fixing graph structure.
line 46
Very specific:
Use prompt-architect when prompts need graph-backed instructions.
line 93
Very specific:
Use design-reviewer when graph findings affect UX quality.
line 94
Very specific:
Use product-designer when a missing object reflects unclear product intent.
line 95
Very specific:
Use accessibility-reviewer when checklists or validation must cover accessibility.
line 96

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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

Knowledge Graph Architect Skill

Role

Maintain AI Design Context as a schema-first knowledge graph.

Mission

Keep objects, metadata, generated indexes, relationships, and validation behavior consistent as the repository grows.

When To Use

A change affects schemas, registry files, generated indexes, object metadata, or relationship types.
A contributor needs to add or migrate a knowledge object type.
The graph has orphans, missing targets, duplicate concepts, or stale generated files.
DesignLint readiness or validation scope is being changed.

When Not To Use

The task is only product UX, visual design, or accessibility review.
The request asks for new design advice without source research.
The change is a local UI implementation with no repository architecture impact.

Inputs

docs/KNOWLEDGE_ENGINE.md
schema/*.schema.json
registry/objects.json
registry/relationships.json
Generated indexes and graph report.
Target files or object type being changed.

Process

Identify affected object types.
Check ID, alias, slug, status, maturity, and relationship conventions.
Verify relationship direction and type.
Check generated indexes and graph diagnostics.
Run or request npm run check.
Flag missing schema, registry, or validation coverage.
Avoid changing content meaning while fixing graph structure.

Output Format

## Graph Assessment
- Object types affected:
- Registry impact:
- Relationship impact:
- Generated index impact:

## Findings
- Issue:
- File:
- Suggested fix:

## Validation
- Command:
- Result:

## Handoff
- Next skill:
- Reason:

Quality Checklist

IDs are stable and globally unique.
Aliases and slugs are unique where required.
Relationship targets resolve.
Generated indexes are not edited manually.
Manual indexes are treated as transitional.
Validation scope is honestly documented.

Related Rules

RULE-00003 / IA-001 One Stable Home Surface
RULE-00004 / IA-002 Define The Object Model Before Screens
RULE-00008 / VIS-001 Semantic Tokens Only

Related Patterns

PAT-00001 / PAT-001 Daily Home Surface
PAT-00006 / PAT-006 Object Status List

Handoff To Other Skills

Use prompt-architect when prompts need graph-backed instructions.
Use design-reviewer when graph findings affect UX quality.
Use product-designer when a missing object reflects unclear product intent.
Use accessibility-reviewer when checklists or validation must cover accessibility.

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