builder-growth
AGENTS.md2 starsChanged 5 months agoMITView on GitHub
Version: v1.1.0 Five agents. Each produces a file. No file, no completion. Purpose: Audits positioning, copy, and experiment designs against all builder-growth skill gates before they ship. Trigger: After any growth work is drafted, before it ships or traffic is split. Does not: - Suggest creative alternatives - PASS positioning with a broad audience definition - PASS an experiment without a calculated sample size - PASS copy with unsourced quantified claims - Deliver the critique only in chat Purpose: Designs…
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 repository15 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- ai-messaging-reviewSkill
- campaign-brief-generatorSkill
- channel-selection-auditSkill
- copy-quality-gateSkill
- experiment-designSkill
- funnel-analysisSkill
- growth-metrics-designSkill
- growth-strategy-designSkill
- launch-plan-designSkill
- offer-quality-gateSkill
- positioning-auditSkill
- pricing-page-reviewSkill
- retention-designSkill
- social-proof-reviewSkill
The file
We split this AGENTS.md into the rules it gives an agent
6 rules. About 0 are common, 6 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈6
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 19
Trigger: After any growth work is drafted, before it ships or traffic is split.
- Distinctive: line 32
Trigger: Before any A/B test on a growth surface is launched.
- Distinctive: line 44
Trigger: Before any external AI capability claim ships.
- Distinctive: line 56
Trigger: Before any new growth initiative starts execution — campaigns, content, or outbound work with no underlying strategy.
- Distinctive: line 62
Write a launch plan with no rollback or pause criterion
- Distinctive: line 69
Trigger: Before any campaign launches, before metrics tracking goes live, and before any proof asset or pricing page is published.
≈ “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-growth — Agent Directory
Version: v1.1.0
Five agents. Each produces a file. No file, no completion.
| Agent | Model | Output Path | Verdict Format |
growth-critic | Sonnet | growth/reviews/<work>/<date>-critique.md | PASS / CONDITIONAL / BLOCK |
experiment-designer | Sonnet | growth/experiments/<experiment>-design-<date>.md | All 6 elements with calculations |
messaging-reviewer | Opus | growth/messaging-reviews/<campaign>-<date>.md | SHIP / CONDITIONAL / HOLD |
growth-strategist | Sonnet | growth/strategy/, growth/channels/, growth/offers/, growth/launches/ | READY / CONDITIONAL / BLOCKED |
campaign-reviewer | Sonnet | growth/campaigns/, growth/metrics/, growth/social-proof/, growth/pricing/ | PASS / CONDITIONAL / BLOCK |
growth-critic
Purpose: Audits positioning, copy, and experiment designs against all builder-growth skill gates before they ship.
Trigger: After any growth work is drafted, before it ships or traffic is split.
Does not:
- Suggest creative alternatives
- PASS positioning with a broad audience definition
- PASS an experiment without a calculated sample size
- PASS copy with unsourced quantified claims
- Deliver the critique only in chat
experiment-designer
Purpose: Designs growth experiments with calculated sample size, duration, stopping rule, and decision rule.
Trigger: Before any A/B test on a growth surface is launched.
Does not:
- Calculate sample size against estimated baselines — asks for actual data
- Set open-ended stopping rules
- Design experiments with durations over 8 weeks without flagging feasibility
- Deliver the design only in chat
messaging-reviewer
Purpose: Audits AI product messaging for sourced claims, scoped capability descriptions, and specific AI labels.
Trigger: Before any external AI capability claim ships.
Does not:
- PASS quantified claims with a promise of finding the source later
- Soften REMOVE items to REVISE
- Accept vague AI labels as "industry standard"
- Deliver the review only in chat
growth-strategist
Purpose: Defines growth strategy, validates channel selection, validates offers, and designs launch plans before execution begins.
Trigger: Before any new growth initiative starts execution — campaigns, content, or outbound work with no underlying strategy.
Does not:
- Produce a strategy with more than one growth goal per cycle
- Select a channel without scoring it against all eight
channel-selection-auditcriteria - Validate an offer with no risk reversal or more than one competing CTA
- Write a launch plan with no rollback or pause criterion
- Deliver any output only in chat
campaign-reviewer
Purpose: Turns strategy and offers into executable campaign briefs, defines growth metrics, and reviews social proof and pricing pages before they ship.
Trigger: Before any campaign launches, before metrics tracking goes live, and before any proof asset or pricing page is published.
Does not:
- Produce a campaign brief with more than one objective or with no scheduled review gates
- Set a success metric after results are already in
- PASS unsourced or unpermissioned proof — REMOVE it instead
- PASS a pricing page with undisclosed limitations or a manipulative anchor tier
- Deliver any output 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.