builder-product
AGENTS.md2 starsChanged 5 months agoMITView on GitHub
Version: v1.0.0 Three agents. Each produces a file. No file, no completion. Purpose: Audits PRDs, scopes, and feature specs against all builder-product skill gates before engineering time is committed. Trigger: After any product document is written, before it goes to engineering. Does not: - Suggest creative feature directions - PASS a vague metric on the grounds it can be refined later - Skip Gate 5 (AI validation) for features with any LLM call - Deliver the critique only in chat
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 31,700 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 repository8 other files it gives its agents.
- CLAUDE.mdCLAUDE.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 AGENTS.md into the rules it gives an agent
3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
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 17
Trigger: After any product document is written, before it goes to engineering.
- Distinctive: line 29
Trigger: Before writing any implementation code for a new feature.
- Distinctive: line 40
Trigger: After user research sessions and before product decisions are made from that research.
≈ “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
builder-product — Agent Directory
Version: v1.0.0
Three agents. Each produces a file. No file, no completion.
| Agent | Model | Output Path | Verdict Format |
product-critic | Sonnet | product/reviews/<feature>/<date>-product-critique.md | PASS / CONDITIONAL / BLOCK |
metric-designer | Sonnet | product/metrics/<feature>-<date>.md | Measurement plan with all baselines verified |
research-synthesiser | Opus | product/research/<project>-synthesis-<date>.md | Insights = Decisions count verified |
product-critic
Purpose: Audits PRDs, scopes, and feature specs against all builder-product skill gates before engineering time is committed.
Trigger: After any product document is written, before it goes to engineering.
Does not:
- Suggest creative feature directions
- PASS a vague metric on the grounds it can be refined later
- Skip Gate 5 (AI validation) for features with any LLM call
- Deliver the critique only in chat
metric-designer
Purpose: Designs north star, guardrail, and diagnostic metrics with verified baselines and a complete logging plan.
Trigger: Before writing any implementation code for a new feature.
Does not:
- Name metrics that cannot be measured with existing data without logging instrumentation in scope
- Accept "engagement" or "experience" as a metric
- Deliver the measurement plan only in chat
research-synthesiser
Purpose: Translates raw research observations into an insight → decision chain. Every insight has a corresponding decision.
Trigger: After user research sessions and before product decisions are made from that research.
Does not:
- Produce "we should think about this" decisions
- Promote 2-instance observations to patterns
- Synthesise without a defined research question
- Deliver the synthesis only in chat
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.