agentleFS
EgiStrEgiStr/deckto/skills/audience-fit/SKILL.md

audience-fit

Skill0 starsChanged 50 days agoMITView on GitHub

Use when a deck's audience must be profiled before tone, depth, font size, or slide count can be decided — called by pitchdeck-pitch-me Phase 1 phase B, or standalone when someone asks "who is this deck for" / "how should I pitch this to X". Produces an audience profile with explicit directives. Do NOT use as a substitute for the full pitch-me questioning framework.

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 2 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

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

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

    Ask one at a time (same discipline as pitchdeck-pitch-me):

    line 23
  2. Very specific:

    Write this block into pitch.md (pitch-me) or return it standalone:

    line 47
  3. Very specific:

    Never invent the audience: If the user doesn't know, that's an unknown entry — an invented audience silently corrupts every downstream …

    line 67
  4. Distinctive:

    Font floor override is rare: Only for genuine room-size constraints (a keynote in a 2,000-seat hall raises floors.

    line 71
  5. Distinctive:

    Default is always the config value — this is an escape hatch, not a dial.

    line 73

≈ “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

Audience Fit — who is in the room, and what that changes

Profile the deck's audience and emit directives that downstream stages obey: tone, depth, evidence type, font size floor, and recommended slide count.

Core principle: An insight is only an insight relative to a specific audience. "so what?" has no answer without a "so what, to whom" — so this runs before grinding, not after.

When to Use

  • pitchdeck-pitch-me Phase 1 (phase B) calls it
  • Standalone: "who should this deck target", "how do I pitch this to <audience>"
  • Re-profiling when the audience changes mid-deck (rare, but it invalidates tone)

Phase 1 — Four questions

Ask one at a time (same discipline as pitchdeck-pitch-me):

#QuestionWhy it changes the deck
1Who exactly is in the room? (roles, not "investors")Role determines what counts as evidence
2What do they already know?Sets the baseline — explaining known things wastes the deck
3What decision can they actually make?Bounds the ask; asking beyond their authority wastes the close
4What's their strongest objection?Becomes the highest-value slide (usually the moat/risk slide)

Phase 2 — Profile

Map to a profile. When someone spans two, pick the narrower — you can always under-explain less.

ProfileOptimize forEvidence that landsTypical length
investorreturn, market, defensibility, teamtraction numbers, unit economics, market sizing10–15 slides
boarddecisions, risk, tradeoffstrend lines, variance vs plan, named risks12–20
technical-clientfeasibility, integration, correctnessarchitecture, benchmarks, honest limits15–25
executivecost, time, risk to themthe recommendation, options, cost of delay5–10
generalclarity, story, one takeawaynarrative, concrete examples, analogy10–20

Phase 3 — Emit directives

Write this block into pitch.md (pitch-me) or return it standalone:

audience_profile:
  profile: "investor"                # from the table
  roles: ["seed-stage partner", "associate"]
  known_baseline: "<what we can skip explaining>"
  decision_authority: "<what they can approve>"
  primary_objection: "<the strongest pushback>"
directives:
  tone: "professional"               # feeds humanizer --voice
  depth: "shallow-broad"             # vs deep-narrow
  evidence_priority: "traction-first"
  font_floor_override: null          # null = use deckto.config.json floors
  recommended_slides: 12
  must_include_slide: "<the objection slide>"

Rules

  • Never invent the audience. If the user doesn't know, that's an unknown entry — an invented audience silently corrupts every downstream choice.
  • The objection is required. An audience with no stated objection has not been profiled; ask again.
  • Font floor override is rare. Only for genuine room-size constraints (a keynote in a 2,000-seat hall raises floors; a 1:1 read-on-laptop deck may lower them). Default is always the config value — this is an escape hatch, not a dial.
  • must_include_slide is the point of this skill. The objection slide is usually the one amateurs cut.

Honesty note

The profile table is [C] practitioner convention (pitch-coaching canon, investor guidance), not empirically validated audience segmentation. Deck-length ranges are conventions that vary wildly by context. Recorded as conventions deliberately — the skill's value is forcing the four questions, not the precision of the table.

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.