agentleFS
gabros20gabros20/product-skill/site/llms.txt

product-skill / site

llms.txt1 starsChanged 4 months agoMITView on GitHub

Decide what to build, for whom, and why — then specify it precisely enough that design and engineering build the right thing. A faceted agent skill for the front of the product lifecycle: discovery, strategy, positioning, business model, market sizing, PRDs and acceptance criteria, prioritization, success metrics, and a living risk register. product owns the decision about what to build, not its execution. SKILL.md is a faceted router: select one primary job (of nine) and at most one surface overlay…

The file

Files
llms.txt
# product

> Decide what to build, for whom, and why — then specify it precisely enough that design and engineering build the right thing. A faceted agent skill for the front of the product lifecycle: discovery, strategy, positioning, business model, market sizing, PRDs and acceptance criteria, prioritization, success metrics, and a living risk register.

product owns the *decision* about what to build, not its execution. SKILL.md is a faceted router: select one primary job (of nine) and at most one surface overlay (of five) that reshapes the artifact for the product type — read only those references fully, never the whole library. Every claim is tagged Fact / Inference / Assumption; numbers carry source-calibrated confidence; prioritization runs every applicable framework and reports the divergence. Agentic/AI is a first-class surface: for a model that acts, the eval set *is* the acceptance criteria, gating launch. Installs as plain Agent Skills format (agentskills.io) on Claude Code, Codex, Cursor, Antigravity, opencode, Grok Build, Hermes, and more.

## Install

- npm/skills.sh: `npx skills add gabros20/product-skill -g -y`
- Clone + installer: `git clone https://github.com/gabros20/product-skill && cd product-skill && ./install.sh codex`
- Manual: `cp -R skills/product ~/.claude/skills/product`

## Docs

- [README.md](https://raw.githubusercontent.com/gabros20/product-skill/main/README.md)
- [skills/product/SKILL.md](https://raw.githubusercontent.com/gabros20/product-skill/main/skills/product/SKILL.md)
- [docs/installation.md](https://raw.githubusercontent.com/gabros20/product-skill/main/docs/installation.md)
- [docs/usage.md](https://raw.githubusercontent.com/gabros20/product-skill/main/docs/usage.md)
- [docs/recipes.md](https://raw.githubusercontent.com/gabros20/product-skill/main/docs/recipes.md)

## Reference inventory (`skills/product/references/`) — 15 files

### Primary jobs (pick exactly one)
- `discovery.md` — validate the problem: Opportunity Solution Tree (Torres), JTBD (Christensen), The Mom Test, MITRE Problem Framing, Proof-of-Life probe
- `strategy.md` — the bet: strategy kernel (Rumelt), strategy pyramid (Perri), product principles, Lean Canvas (Maurya), Foundation Sprint
- `positioning.md` — the frame: Moore positioning statement & beachhead, Dunford's method, Value Proposition Canvas (Osterwalder)
- `business-model.md` — Business Model Canvas + pricing hypothesis: value-based method, Van Westendorp, tier psychology
- `market.md` — TAM/SAM/SOM (top-down/bottom-up/comparable) with confidence tiers; Porter's Five Forces / SWOT / PESTEL; competitor profiles
- `prd.md` — altitude-flexed PRD (one-pager ↔ full ↔ machine-readable ProductSpec), Given/When/Then acceptance criteria, job/user stories
- `prioritization.md` — multi-engine: RICE, ICE, Opportunity Score, Kano, MoSCoW, Weighted Scoring; divergence is the finding; outcome roadmap
- `metrics.md` — North-star + metric tree, HEART (Google), AARRR (McClure), Lean Analytics criteria, guardrails, OKRs (Doerr)
- `risk-register.md` — living register: 8-category taxonomy (Cagan), demand hierarchy L0–L4, evidence ladder, four-verdict contract, red-team, pre-mortem

### Surface overlays (at most one)
- `surface-website.md` — marketing/content: conversion, proof strategy, Attention Ratio
- `surface-app-saas.md` — accounts & recurring usage: activation/retention, pricing tiers, MRR/churn
- `surface-mobile.md` — store distribution, ASO, permissions, interruption-driven sessions
- `surface-agentic.md` — **the differentiator**: capability & failure-mode mapping, eval-as-acceptance-criteria, human-in-the-loop autonomy, non-determinism/latency/cost/safety risk, model choice
- `surface-internal.md` — internal/enterprise tools: adoption over acquisition, buyer ≠ user, cost-based prioritization

### Pipeline
- `handoff.md` — standalone vs pipeline behavior and the `handoff.yaml` companion consumed by `design` and `architecture`

## Optional

- [Visual guide](https://product-skill.vercel.app): the faceted router, the nine jobs, five surfaces, the agentic wedge, invariants, and the handoff seam on one page.

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.