builder-product
CLAUDE.md2 starsChanged 5 months agoMITView on GitHub
builder-product — AI Product Skills Pack v1.0.0 For any new feature: Standard product frameworks produce documents. This pack produces decisions: - Every PRD has a measurable success metric with a baseline - Every estimate references a written scope - Every research report has an insight → decision chain - Every AI feature addresses hallucination UX, trust calibration, and reversibility A completed gate means a decision was made. Not a document was written.
In this repository8 other files it gives its agents.
- AGENTS.mdAGENTS.md
- ab-test-designSkill
- ai-feature-validationSkill
- feature-scopingSkill
- metric-definitionSkill
- mvp-scope-validatorSkill
- prd-quality-gateSkill
- user-research-synthesisSkill
The file
We split this CLAUDE.md into the rules it gives an agent
4 rules. About 0 are common, 4 are distinctive, and 0 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.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 42
Every PRD has a measurable success metric with a baseline
- Distinctive: line 43
Every estimate references a written scope
- Distinctive: line 44
Every research report has an insight → decision chain
- Distinctive: line 45
Every AI feature addresses hallucination UX, trust calibration, and reversibility
≈ “Common” is an estimate: at least 1 in 5 of the 26,980 CLAUDE.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
builder-product — Claude Configuration
Pack Version
builder-product — AI Product Skills Pack v1.0.0
File Conventions
| Artifact | Path |
| PRDs | product/prd/<feature>-<date>.md |
| Feature scopes | product/scopes/<feature>-<date>.md |
| Metric plans | product/metrics/<feature>-<date>.md |
| Research synthesis | product/research/<project>-synthesis-<date>.md |
| Experiment designs | product/experiments/<feature>-design-<date>.md |
| AI feature validations | product/ai-validation/<feature>-<date>.md |
| Product critiques | product/reviews/<feature>/<date>-product-critique.md |
Skill Trigger Sequence
For any new feature:
mvp-scope-validator→ before any PRD, when the idea may be over-scoped — extract hypothesis and define minimum scopeprd-quality-gate→ defines the problem and success metricfeature-scoping→ written scope before any estimatemetric-definition→ measurement plan with baselines before codeai-feature-validation→ required if any LLM call is involveduser-research-synthesis→ if any research was done, synthesise before using itab-test-design→ before any experiment that informs a ship/no-ship decision
Agent Usage
| Agent | When to Use |
product-critic | After writing any PRD or feature spec, before engineering time is committed |
metric-designer | When designing measurement frameworks for new features |
research-synthesiser | After user research sessions, before decisions are made |
What Makes This Pack Different From Generic Product Frameworks
Standard product frameworks produce documents. This pack produces decisions:
- Every PRD has a measurable success metric with a baseline
- Every estimate references a written scope
- Every research report has an insight → decision chain
- Every AI feature addresses hallucination UX, trust calibration, and reversibility
A completed gate means a decision was made. Not a document was written.
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.