agentleFS
jqaisystemsjqaisystems/jqai-ai-skills/skills/write-evidence-backed-profile/SKILL.md

write-evidence-backed-profile

Skill3 starsChanged 6 months agoMITView on GitHub

Draft bios, applications, portfolio introductions, proposal introductions, cover notes, and professional profile copy from approved evidence without inventing experience or results. Use when a user wants personal or company positioning tailored to an audience, role, opportunity, directory, portfolio, proposal, or public profile and factual grounding matters.

The file

Show its rules and text

We split this skill into the rules it gives an agent

13 rules. About 0 are common, 13 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈13

    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
  1. Distinctive:

    Create audience-specific profile writing from an approved evidence set.

    line 8
  2. Distinctive:

    Read references/channel-output-rules.md for the selected format.

    line 14
  3. Distinctive:

    Use assets/evidence-pack-template.md when the user has not supplied a structured source set.

    line 19
  4. Distinctive:

    Use public sources and user-approved facts only.

    line 20
  5. Distinctive:

    Verify current pages when roles, availability, metrics, projects, credentials, services, or positioning may have changed.

    line 21
  6. Distinctive:

    Treat inaccessible sources and unsupported user expectations as unverified.

    line 22
  7. Distinctive:

    Do not embed a specific person's identity, CV, website, source registry, or voice guide in the skill.

    line 23
  8. Distinctive:

    Do not infer job titles, customer relationships, years of experience, technical depth, revenue, performance, availability, awards, or …

    line 34
  9. Distinctive:

    Use concrete public work when it improves credibility.

    line 40
  10. Distinctive:

    Avoid inflated certainty, unsupported superlatives, fabricated enthusiasm, and irrelevant capability lists.

    line 42
  11. Distinctive:

    Verify every name, title, date, link, metric, project description, and factual statement used.

    line 47
  12. Distinctive:

    Keep sensitive or user-approved non-public material out of reusable examples and public artifacts.

    line 49
  13. Distinctive:

    Never submit a form, send a message, edit a public profile, or publish the draft.

    line 50

≈ “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

Write Evidence-Backed Profile

Create audience-specific profile writing from an approved evidence set. Prefer a smaller defensible story over a broad invented persona.

1. Define the writing task

  • Confirm the audience, channel, prompt or field, word or character limit, desired action, and tone.
  • Identify the claims the user expects the draft to communicate.
  • Read references/channel-output-rules.md for the selected format.
  • Preserve exact names, links, titles, dates, and wording supplied for the current task.

2. Build or inspect the evidence pack

  • Use assets/evidence-pack-template.md when the user has not supplied a structured source set.
  • Use public sources and user-approved facts only.
  • Verify current pages when roles, availability, metrics, projects, credentials, services, or positioning may have changed.
  • Treat inaccessible sources and unsupported user expectations as unverified.
  • Do not embed a specific person's identity, CV, website, source registry, or voice guide in the skill.

3. Map evidence to the audience

For each important requested claim, record:

  • the evidence that supports it
  • whether it is direct experience, adjacent experience, interpretation, or unsupported
  • why it matters to this audience
  • the strongest wording the evidence permits

Do not infer job titles, customer relationships, years of experience, technical depth, revenue, performance, availability, awards, or credentials from weak signals.

4. Write the strongest honest version

  • Lead with the most relevant supported overlap between the evidence and the audience's need.
  • Present adjacent experience as adjacent, not direct expertise.
  • Use concrete public work when it improves credibility.
  • Omit unsupported claims instead of filling gaps with generic confidence.
  • Avoid inflated certainty, unsupported superlatives, fabricated enthusiasm, and irrelevant capability lists.
  • Match the requested format so the result is ready for human review and use.

5. Complete the evidence check

  • Verify every name, title, date, link, metric, project description, and factual statement used.
  • List requested claims that were qualified or omitted and explain the evidence gap briefly.
  • Keep sensitive or user-approved non-public material out of reusable examples and public artifacts.
  • Never submit a form, send a message, edit a public profile, or publish the draft.

Return the compact evidence map, finished draft, qualifications, omitted claims, sources checked, and any fact that still needs confirmation.

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.