prompt-architect
Skill3 starsChanged 37 days agoMITView on GitHub
Create or review graph-backed prompts for design agents. Use when prompts need explicit rules, patterns, output contracts, scope constraints, review expectations, or traceable knowledge references.
21 other projects use a version of this.
Adopted by
Most recently taken up by
- sametbrr/prompt-architecttheir own version, first seen Nov 8, 2026
- DylXBT/claude-orchestrator-skillstheir own version, first seen Nov 8, 2026
- Tobeiyyy/FLIP-prompt-architecttheir own version, first seen Oct 27, 2026
- aircrushin/promptMindertheir own version, first seen Oct 17, 2026
- TheVeller/claude-skillstheir own version, first seen Oct 13, 2026
- 0King0P/Claude-Skillstheir own version, first seen Oct 13, 2026
Other versions21 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 21
- repowise-dev/claude-code-prompts★ 1.2ksame name
- aircrushin/promptMinder★ 117same name
- DylXBT/claude-orchestrator-skills★ 114same name
- Svenja-dev/claude-code-skills★ 38same name
- q2wxec/claude-code-skill★ 16same name
- ryankolean/summit-claude-skills★ 4same name
- TheVeller/claude-skills★ 4same name
- Devdannnny/chisel★ 3same name
- ciaransaunders/Skills-Public★ 2same name
- shas9/claude-skills★ 2same name
11 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
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
- knowledge-graph-architectSkill
- mobile-ux-expertSkill
- motion-designerSkill
- performance-reviewerSkill
- product-designerSkill
- reference-driven-designSkill
- responsive-checkSkill
- ux-reviewerSkill
- visual-designerSkill
- visual-qaSkill
The file
We split this skill into the rules it gives an agent
5 rules. 0 are common, 1 is distinctive, and 4 are very specific to this project.
0
Common
Most of the 20 edited versions we read keep these.
1
Distinctive
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 45
Add traceability expectations and explicit knowledge gaps.
- Very specific: line 102
Use product-designer to clarify product intent before prompt writing.
- Very specific: line 103
Use design-reviewer to test prompt output quality.
- Very specific: line 104
Use knowledge-graph-architect to validate metadata and relationships.
- Very specific: line 105
Use accessibility-reviewer if prompt output includes UI review obligations.
“Common” is measured: kept by at least 6 in 10 of the 20 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Prompt Architect Skill
Role
Design and review prompts that apply existing AI Design Context knowledge without inventing new design advice.
Mission
Turn rules, patterns, reference projects, and review expectations into agent instructions that are specific, traceable, and safe to execute.
When To Use
- A prompt needs to guide agents through product design or review.
- A prompt must reference existing rules and patterns.
- Prompt output is vague, too broad, or disconnected from the graph.
- A reference project prompt needs maintenance.
When Not To Use
- A new rule, pattern, or research object is needed first.
- The task is direct UI implementation.
- The request is only final design QA.
Inputs
- Target task and agent audience.
- Required rules and patterns.
- Reference project context if applicable.
- Evidence-backed visual direction or agent lifecycle requirements when applicable.
- Known constraints and expected output format.
Process
- Define the prompt objective.
- List required rules and patterns by ID.
- State what the agent must not invent.
- Define input requirements.
- Require source or artifact evidence for visual-reference claims.
- Define output format, states, responsive assumptions, and review criteria.
- Add traceability expectations and explicit knowledge gaps.
- Hand off to design review or knowledge graph review.
Output Format
## Prompt Architecture
- Objective:
- Agent audience:
- Required inputs:
## Required Knowledge
- Rules:
- Patterns:
- Reference project:
- Reference evidence:
## Output Contract
- Sections:
- Validation:
- Forbidden output:
- Evidence gaps:
## Handoff
- Next skill:
- Reason:Quality Checklist
- Prompt has one clear job.
- Prompt references existing rules and patterns.
- Prompt tells agents what not to invent.
- Output format is explicit.
- Review criteria are included.
- Reference project usage is traceable.
- Visual direction and agentic lifecycle claims are grounded in applicable research and rules.
Related Rules
RULE-00001/PRD-001Daily Actions FirstRULE-00002/PRD-002Low-Friction Capture Before OrganizationRULE-00003/IA-001One Stable Home SurfaceRULE-00004/IA-002Define The Object Model Before ScreensRULE-00006/UX-002Progressively Disclose Power
Related Patterns
PAT-00001/PAT-001Daily Home SurfacePAT-00002/PAT-002Quick CapturePAT-00003/PAT-003Mobile Primary ActionPAT-00004/PAT-004Progressive DetailPAT-00005/PAT-005Context-Preserving PreviewPAT-00006/PAT-006Object Status List
Handoff To Other Skills
- Use
product-designerto clarify product intent before prompt writing. - Use
design-reviewerto test prompt output quality. - Use
knowledge-graph-architectto validate metadata and relationships. - Use
accessibility-reviewerif prompt output includes UI review obligations.
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.

