product
Skill0 starsChanged 4 months agoMITView on GitHub
Use when creating, amending, or editing `PRODUCT.md`. Owns the foundational product-context document — Vision, Core Value Proposition, Business Model, Business Goals, Target Users, Product Boundaries, Strategic Direction.
16 other projects use a version of this.
Adopted by
Most recently taken up by
- mahirhacks/mstacktheir own version, first seen Nov 16, 2026
- fatihkan/baditheir own version, first seen Nov 16, 2026
- Mark-Life/agent-skillstheir own version, first seen Nov 10, 2026
- Astatide1337/Skillstheir own version, first seen Nov 8, 2026
- PeterPartridge/CooperativeAICodingtheir own version, first seen Nov 8, 2026
- gabros20/product-skilltheir own version, first seen Nov 1, 2026
Used with
Projects with it also use review and research.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
SentrySkillreview
Fixture review skill.
Also in 3 projects that use this
NVIDIASkillresearch
Before designing a feature, research how comparable libraries solve the problem, check best practices, known pitfalls, and upstream changes. Use when starting a new feature, investigating an unfamiliar area, or the user asks to research something.
Also in 3 projects that use this
Other versions16 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 16
- nitinjain999/platform-skills★ 71same name
- fatihkan/badi★ 8same name
- malinskibeniamin/skills★ 4same name
- Mark-Life/agent-skills★ 4same name
- websublime/mister-anderson★ 3same name
- alex-voloshin-dev/ai-skills★ 2same name
- PeterPartridge/CooperativeAICoding★ 2same name
- tikhomirovv/agents-knowledge★ 2same name
- Astatide1337/Skills★ 1same name
- donatassimkus/claude-ai-skills★ 1same name
6 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 repository8 other files it gives its agents.
- checkpointSkill
- design-themeSkill
- dispatch-agentsSkill
- githubSkill
- github-templatesSkill
- gitSkill
- technical-designSkill
- user-storiesSkill
The file
We split this skill into the rules it gives an agent
7 rules. 0 are common, 2 are distinctive, and 5 are very specific to this project.
0
Common
Most of the 14 edited versions we read keep these.
2
Distinctive
This author's own choices. Worth a look.
5
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 33
Read PRODUCT.md (absent → stop: `No PRODUCT.md found.
- Very specific: line 57
After three failed follow-ups, offer to mark the section Not yet defined rather than write something vague.
- Distinctive: line 127
Create asks each verbatim.
- Distinctive: line 152
Never invent content — every body comes from interview answers or confirmed README-seeded suggestions.
- Very specific: line 153
Never write placeholder braces ({...}) or "TBD" into the final file.
- Very specific: line 154
Never silently overwrite an existing PRODUCT.md — always confirm.
- Very specific: line 155
Never write skill-adjacent files (ROADMAP.md, etc.) — surface the suggestion, let the caller route.
“Common” is measured: kept by at least 6 in 10 of the 14 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
PRODUCT.md Authoring
Mode Detection
PRODUCT.md present? | Intent verb in the request | Mode |
| no | create / set up / generate / init / draft | Create |
| yes | create / generate / regenerate | Confirm: Skip (default) or Regenerate |
| yes | rewrite / overhaul / start over | Confirm: Skip (default) or Overwrite |
| yes | amend / update / change / edit / revise | Amend |
| yes | read-only ("what's our vision?", "show me the boundaries") | Summarise; do not modify |
Both destructive ops (Regenerate, Overwrite) require confirmation; Skip is the default.
Create
Confirm PRODUCT.md absent (or Regenerate/Overwrite chosen). Then:
- Interview through the Question Bank in canonical section order — one
AskUserQuestioneach, applying Specificity Tests with follow-ups. - Surface any out-of-canon content (offer a bottom section or omit).
- Consistency pass over all answers → resolve contradictions.
- Author
PRODUCT.mdat the repo root per File Structure, concrete content. - Confirm with path + one-line summary per section.
Amend
Read PRODUCT.md (absent → stop: No `PRODUCT.md` found. Create it first.). Capture the full baseline verbatim — every section, subsection, non-canonical content; non-targeted sections preserved byte-for-byte. Determine target section(s) from the arg via Section Argument Map, else ask which. Offer cascade revisits (not forced):
- Vision → Core Value Proposition, Strategic Direction
- Target Users → Core Value Proposition, Product Boundaries
- Business Model → Business Goals, Strategic Direction
Interview targeted section(s) → apply Specificity Tests → consistency pass against untouched baseline → rewrite per Apply Rules → overwrite PRODUCT.md → confirm section(s) updated.
Specificity Tests
Apply after every interview answer; on failure ask a focused follow-up.
General (all sections): could a new hire read this section alone and act consistently with it? If they'd ask "what did they really mean?", it fails.
| Section | Test |
| Vision | Names a specific user, problem, and outcome. "Help people be more productive" fails; "Help engineering managers run 1:1s that surface blockers earlier" passes. |
| Core Value Proposition | Names the alternative being chosen against. "Faster checkout" fails; "Faster checkout than Shopify's default for stores with >10k SKUs" passes. |
| Business Model | Names who pays, what for, and roughly how much. Vague free-tier descriptions fail. |
| Business Goals | Each goal measurable or observable. "Improve user experience" fails; "Reduce time-to-first-value from 12 minutes to under 5" passes. |
| Target Users | Primary user is a single role with named pain points, not a demographic. "Millennials" fails; "Solo founders in their first year, juggling sales and product" passes. |
| Product Boundaries | In/Out of Scope each have a concrete example. "Various integrations" fails; "Stripe, Slack, Linear at launch — no Salesforce, no Microsoft stack" passes. |
| Strategic Direction | Names 1–2 priorities and one explicit non-priority. "Grow the platform" fails; "Q3: depth in analytics over breadth of integrations. Not chasing mobile parity this year." passes. |
After three failed follow-ups, offer to mark the section Not yet defined rather than write something vague.
Contract
File Structure
Pure markdown — no frontmatter, no version field. Seven canonical sections in immutable order:
# Product Context
## Vision
{vision statement — one clear paragraph}
## Core Value Proposition
{single clearest reason a customer chooses this product over alternatives — one sentence or short paragraph}
## Business Model
{how the product makes money — one short paragraph}
## Business Goals
{each goal as a bullet point}
## Target Users
### Primary: {primary persona name or role}
{description of role, key needs, and pain points}
### Secondary: {secondary persona name or role}
{description — omit this subsection entirely if no secondary users were identified}
## Product Boundaries
### In Scope
{each in-scope capability as a bullet point}
### Out of Scope
{each exclusion as a bullet point — or "Not specified" if none given}
## Strategic Direction
{strategic direction as a paragraph; name 1–2 priorities and at least one explicit non-priority}Non-canonical user-added sections, if any, appear at the bottom and are preserved untouched on amend.
Section Argument Map
| Argument | Section |
vision | Vision |
value or proposition | Core Value Proposition |
model | Business Model |
goals | Business Goals |
users | Target Users |
boundaries | Product Boundaries |
strategy or direction | Strategic Direction |
all | All sections |
Comma-separated list (e.g. vision,value) | Multiple sections |
Interview Question Bank
Create asks each verbatim; Amend prefixes "the updated" or "now" where natural.
| Section | Question |
| Vision | "What is the product vision? Describe the problem it solves, for whom, and the core value it delivers." |
| Core Value Proposition | "What is the core value proposition? The single clearest reason a customer chooses this product over alternatives — name the alternative." |
| Business Model | "What is the business model? How does the product make money — who pays, what for, and roughly what's the price shape?" |
| Business Goals | "What are the 3–5 key business goals? List them as short, action-oriented, measurable statements (e.g. 'Grow monthly active users to 50k', 'Reduce average checkout time below 30s')." |
| Target Users | "Who are the target users? Describe the primary user as a specific role with named pain points. Include a secondary user if applicable." |
| Product Boundaries | "What is explicitly in scope for this product at launch? What is explicitly out of scope — even if it might seem like an obvious fit?" |
| Strategic Direction | "What is the current strategic direction? Name the 1–2 priorities for the next quarter or two, and at least one explicit non-priority — something you're choosing not to pursue." |
Apply Rules
- Create writes the file end-to-end via File Structure. Every section gets concrete content — no placeholders,
{…}braces, "TBD". - Amend edits only targeted section(s); non-targeted sections preserved byte-for-byte (headings, paragraphs, bullets, subsections, bottom non-canonical sections). Target
all→ rewrite all seven canonical sections, preserve non-canonical. Single section → rewrite only its body (+ subsections), leave heading unchanged. - Specificity Test failing after three follow-ups + user agrees → write
Not yet defined.as the section body.
Constraints
- Section order is immutable — seven sections in the File Structure order.
- Section headings match canonical names exactly.
### Secondary:omitted entirely when no secondary user exists — no empty heading.### Out of Scopefalls back toNot specifiedonly when asked and explicitly given none.PRODUCT.mdis pure markdown — no YAML frontmatter,versionfield, or metadata block.- Never invent content — every body comes from interview answers or confirmed README-seeded suggestions.
- Never write placeholder braces (
{...}) or "TBD" into the final file. - Never silently overwrite an existing
PRODUCT.md— always confirm. - Never write skill-adjacent files (
ROADMAP.md, etc.) — surface the suggestion, let the caller route. - Filesystem: write only
PRODUCT.md.
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.

