agentleFS
TomSzenessyTomSzenessy/AI_TEMPLATE/.agents/skills/pitch-from-evidence/SKILL.md

pitch-from-evidence

Skill0 starsChanged 11 days agoMITView on GitHub

Build a presentation, pitch, demo script, or report whose every number and claim has a source; preparing a talk, stakeholder update, investor or grant pitch, release note, paper, or results write-up

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 4 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Audience and format: Write down who reads or listens, what they decide or care about, and the time or length available.

    line 18
  2. Very specific:

    Attribute gains: When showing improvement, name which step moved the number and by how much, measured the same way (one attempt against one …

    line 24
  3. Distinctive:

    Caveats and questions: Write the weaknesses a skeptic would raise and an honest answer to each, plus the likely questions with short …

    line 27
  4. Distinctive:

    Do-not-claim list: Write what the work does not show (untested cases, proxies, results of others) so nobody says it under pressure.

    line 29
  5. Distinctive:

    Every number in the deliverable appears in the claim table with a source that a teammate re-checked, and the caveat, question, and …

    line 41

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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

Pitch from evidence

Outcome: a pitch, demo script, or results write-up in which every number and claim can be traced to a file, a run, or a commit, and every weakness has an honest answer ready. Load this before preparing a talk, a demo, a stakeholder or client update, an investor or grant pitch, release notes, or a results report. The evidence rules live in docs/verification.md; this skill is the order of work.

Steps

  1. Audience and format. Write down who reads or listens, what they decide or care about, and the time or length available. The order follows from what they need first, not from the order the work was done in.
  2. Claim table first, slides second. List every claim with its source: a file and line, a recorded run, or a commit. A number with no source is cut or marked as an estimate. Opinions are labelled as opinions.
  3. Attribute gains. When showing improvement, name which step moved the number and by how much, measured the same way (one attempt against one attempt; experiments.md).
  4. Caveats and questions. Write the weaknesses a skeptic would raise and an honest answer to each, plus the likely questions with short answers.
  5. Do-not-claim list. Write what the work does not show (untested cases, proxies, results of others) so nobody says it under pressure.
  6. Recount last. Re-read every live number (scores, counts, dates) from its source right before presenting; numbers drift while work continues.

An explorable architecture or flow diagram for the deck: Archify (make upstream ID=archify-diagrams, an agent skill that writes interactive HTML/SVG diagrams), pulled only when a static image is not enough; vet any of its scripts with skill-scout before running them.

Done when

  • Every number in the deliverable appears in the claim table with a source that a teammate re-checked, and the caveat, question, and do-not-claim lists exist next to it.

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.