agentleFS
product-on-purposeproduct-on-purpose/pm-skills/skills/foundation-persona/SKILL.md

foundation-persona

Skill777 starsChanged 10 months agoApache-2.0View on GitHub

Generates an evidence-calibrated product or marketing persona using the canonical v2.5 output contract. Use when shaping artifact perspective, stress-testing decisions, or framing product and GTM strategy.

The file

Show its rules and text

We split this skill into the rules it gives an agent

17 rules. About 0 are common, 10 are distinctive, and 7 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈10

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 7

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    If the user asks for agent, ask them to choose product or marketing.

    line 25
  2. Distinctive:

    Before drafting PM or GTM artifacts that need a clear persona viewpoint

    line 29
  3. Distinctive:

    When teams disagree on priorities and need behavior-grounded tradeoff framing

    line 30
  4. Distinctive:

    When assumptions and confidence levels must be explicit for decision review

    line 31
  5. Distinctive:

    When tailoring downstream work (PRD, stories, launch, messaging, enablement) to a specific user or buyer profile

    line 32
  6. Distinctive:

    No evidence exists at all and a real decision rides on the persona: gather research first.

    line 39
  7. Distinctive:

    When asked to generate a persona, follow these steps:

    line 43
  8. Very specific:

    If mode is omitted, ask for mode selection.

    line 47
  9. Very specific:

    If execution must continue without reply, default to product and state that fallback explicitly.

    line 48
  10. Distinctive:

    Use user-provided context first (goals, audience, domain, constraints, sources).

    line 51
  11. Distinctive:

    If evidence is thin, continue generation but mark gaps and calibrate confidence.

    line 52
  12. Very specific:

    Use references/TEMPLATE.md and choose exactly one of:

    line 55
  13. Very specific:

    Use High|Medium|Low confidence with rationale.

    line 72
  14. Very specific:

    Remove template guidance blockquotes (> notes) from the final output.

    line 77
  15. Distinctive:

    Ensure narrative entries are concrete and decision-changing, not placeholder bullets.

    line 78
  16. Very specific:

    Use one mode only (Product or Marketing) per output.

    line 82
  17. Distinctive:

    Keep section numbering and headings from the selected template.

    line 83

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md
<!-- PM-Skills | https://github.com/product-on-purpose/pm-skills | Apache 2.0 -->

Persona Builder

This skill produces decision-usable personas from one canonical template pack.

Supported Modes

  • product
  • marketing
  • buyer as input alias for marketing (output remains labeled Marketing)

Generated agent mode is out of scope for v2.5.0. If the user asks for agent, ask them to choose product or marketing.

When to Use

  • Before drafting PM or GTM artifacts that need a clear persona viewpoint
  • When teams disagree on priorities and need behavior-grounded tradeoff framing
  • When assumptions and confidence levels must be explicit for decision review
  • When tailoring downstream work (PRD, stories, launch, messaging, enablement) to a specific user or buyer profile

When NOT to Use

  • You need the job context rather than the person -> use define-jtbd-canvas; the canvas captures what customers hire products to do, the persona captures who they are
  • You are mapping internal stakeholders, not customers -> use discover-stakeholder-summary
  • You have raw interviews to synthesize first -> use discover-interview-synthesis; a persona built on unsynthesized notes inherits their noise
  • No evidence exists at all and a real decision rides on the persona: gather research first; the skill labels assumptions honestly but cannot substitute for evidence

Instructions

When asked to generate a persona, follow these steps:

  1. Resolve mode and intent Determine whether the request is product or marketing (buyer alias allowed). If mode is omitted, ask for mode selection. If execution must continue without reply, default to product and state that fallback explicitly.
  2. Collect context and evidence Use user-provided context first (goals, audience, domain, constraints, sources). If evidence is thin, continue generation but mark gaps and calibrate confidence.
  3. Select exactly one template Use references/TEMPLATE.md and choose exactly one of:
    • Product Persona Template
    • Marketing Persona Template
  4. Generate a complete artifact Fill the selected template end-to-end:
    • header + one-sentence core-reality statement
    • metadata table
    • Persona Card
    • sections 1 through 11
    • Evidence & Confidence
  5. Enforce mode boundaries
    • Product mode: focus on workflow behavior, decision patterns, friction, quality bar, and product tradeoffs.
    • Marketing mode: focus on buying triggers, evaluation criteria, committee dynamics, objections, messaging, and GTM implications.
  6. Apply evidence and confidence policy
    • Use High|Medium|Low confidence with rationale.
    • Distinguish validated evidence from assumptions.
    • State open questions and governance follow-up.
  7. Finalize for direct use Remove template guidance blockquotes (> notes) from the final output. Ensure narrative entries are concrete and decision-changing, not placeholder bullets.

Output Contract (v2.5.0)

  • Use one mode only (Product or Marketing) per output.
  • Keep section numbering and headings from the selected template.
  • Preserve the evidence table plus validated/assumed/open-questions/governance blocks.

Quality Checklist

Before finalizing, verify:

  • Exactly one mode is used and clearly labeled
  • buyer inputs are normalized to Marketing
  • Header, core-reality statement, metadata table, and Persona Card are present
  • All 1 through 11 sections from the selected template are present and complete
  • Includes/not-valid boundaries are explicit in the metadata and narrative
  • Evidence table is populated with concrete sources
  • Confidence is High, Medium, or Low with rationale
  • Validated, Assumed, Open questions, and Governance blocks are present
  • Template authoring notes (> guidance lines) are removed from the completed output

Examples

See references/EXAMPLE.md for a completed sample output.

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.