agentleFS
samulee003samulee003/agentic-ppt-skills/skills/presentation-product-grill/SKILL.md

presentation-product-grill

Skill0 starsChanged 24 days agoMITView on GitHub

Turns a source inventory into a presentation brief by resolving audience, problem, outcome, promise, constraints, and success decisions through round-based product grilling. Use when the product-grill gate is current for a presentation project or when a router asks for the presentation brief and consequential decision log.

The file

We split this skill into the rules it gives an agent

9 rules. About 0 are common, 4 are distinctive, and 5 are 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.

  • 5

    Very specific

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

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

    Use when product-grill is the current gate.

    line 8
  2. Very specific:

    Read repo instructions, SOURCE-INVENTORY.md, and the cited source files before asking anything.

    line 10
  3. Very specific:

    Before the first question, follow ../make-presentation/references/sharpen-playback.md: sharpen the task, write a one-paragraph ## Playback …

    line 14
  4. Very specific:

    Write PRESENTATION-BRIEF.md from templates/PRESENTATION-BRIEF.md.

    line 21
  5. Distinctive:

    Update the brief as soon as a question is resolved.

    line 23
  6. Distinctive:

    Keep the brief decision-oriented and concrete.

    line 24
  7. Distinctive:

    Prefer short sentences, explicit tradeoffs, and measurable success tests.

    line 25
  8. Very specific:

    Write only consequential decisions to GRILL-LOG.md from templates/GRILL-LOG.md.

    line 27
  9. Distinctive:

    Stop once the brief is coherent enough for the evidence or story gate to proceed.

    line 33

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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

Grill the presentation product

Use when product-grill is the current gate.

  • Read repo instructions, SOURCE-INVENTORY.md, and the cited source files before asking anything.
  • Investigate discoverable answers in the sources instead of asking the user.
  • Cover the full product model: audience, problem, outcome, promise, evidence threshold, constraints, visual direction & rendering DNA, success test, and out of scope.
  • Mandatory Visual Theme Grilling: When visual direction or rendering style is unconfirmed, ask one focused question with concrete theme archetypes from themes/*.md or benchmark decks (e.g. sticker-pop / macau-wellbeing-sketch for physical sketch with 3px ink & hard shadows, bright-sans for Swiss minimal, aurora for dark glassmorphism). Never allow vague adjectives (e.g. "warm", "clean", "modern") to pass without binding them to a concrete rendering contract (border, shadow, typography scale).
  • Before the first question, follow ../make-presentation/references/sharpen-playback.md: sharpen the task, write a one-paragraph ## Playback at the top of the brief, and proceed on recommended defaults (recorded as Assumed) for reversible choices.
  • Grill in rounds per ../make-presentation/references/grilling-rounds.md when a consequential decision is still unresolved and the sources cannot answer it: ask the whole frontier (every decision whose prerequisites are settled) in one round, numbered, each with a recommended answer worded so "yes" accepts it. Never ask a question that depends on another open one in the same round. Honor "ask one question at a time" if the user requests it.
  • Include a recommended answer with each question, worded so "yes" accepts it.
  • Explain what the decision would invalidate downstream when that impact exists.
  • Use references/autonomous-brief.md for the autonomous fallback when the user is unavailable (headless run) or the remaining product decisions are resolvable from the sources — resolve each open field with a defensible value, its rationale, and a confidence level (defaulting visual direction to an explicit registered theme from themes/*.md rather than generic unstyled UI), and flag any low-confidence load-bearing decision for later human review instead of silently guessing.
  • Let any consequential change to Audience, Problem, Outcome, Promise, Evidence threshold, Constraints, Visual direction, Success test, or Out of scope invalidate downstream gates.

Write PRESENTATION-BRIEF.md from templates/PRESENTATION-BRIEF.md.

  • Update the brief as soon as a question is resolved.
  • Keep the brief decision-oriented and concrete.
  • Prefer short sentences, explicit tradeoffs, and measurable success tests.

Write only consequential decisions to GRILL-LOG.md from templates/GRILL-LOG.md.

  • Append new decisions without deleting or rewriting entries from other grill phases or earlier runs.
  • Preserve existing decisions from story or stage grilling.
  • Log decisions that materially change evidence, story, prototype, stage, or production work.
  • Skip minor wording preferences and reversible drafting details.
  • Stop once the brief is coherent enough for the evidence or story gate to proceed.

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.