skill
executive-presence-framework
Two Claude Skills for high-stakes workplace communication, built on 12 frameworks (BLUF, Minto, PMI, SCR, A·D·A). One writes the words you need; one drills you until you can.
Set up
Skills are installed by placing their folder where your agent loads skills.
The download comes from github.com/yshameer/executive-presence-framework itself, not from our server, so what you get is exactly what its author published.
Install instructions come from the README
README
synced Oct 23, 2026Executive Presence Framework
yshameer.github.io/executive-presence-framework
Two agent skills for high-stakes workplace communication, built on 12 frameworks. Works with Claude Code and Codex — both read the same SKILL.md format, so one set of files serves both.
| Skill | What it does | Use when |
|---|---|---|
executive-presence | Diagnoses the moment, picks the frameworks, and writes the actual words you'd say or send | You have a real situation to handle in the next hour |
executive-presence-practice | Gives you a scenario, waits for your answer, then critiques it | You want to get better, not get a script |
The split matters. The coach writes for you; the drill partner refuses to. A drill where the model hands you a model answer up front is just coaching with extra steps.
Both skills share the same 12-framework reference — BLUF, Minto Pyramid, C·D·S, PMI, SCR, Story Structure, Decisive Answering, Meeting Lead, A·D·A, Read Before You Respond, Pause Principle, Consistency Compound.
Install
The script installs for whichever agents you have, and takes --claude, --codex, --project or --dry-run:
git clone https://github.com/yshameer/executive-presence-framework.git
cd executive-presence-framework && ./install.sh
Or copy the folders yourself — a skill is just a directory:
| Agent | Personal | Project |
|---|---|---|
| Claude Code | ~/.claude/skills/ | <repo>/.claude/skills/ |
| Codex | ~/.agents/skills/ | <repo>/.agents/skills/ |
Claude Code can also install this repo as a plugin, which keeps it updatable:
/plugin marketplace add yshameer/executive-presence-framework
Using neither agent? Paste a self-contained prompt from prompts/ into any chat interface.
Full detail for every path in docs/INSTALL.md.
Use
Describe your situation and the right skill triggers itself — on both agents:
"I have to tell leadership on Thursday that the migration is slipping by a month."
→ coach: diagnosis, frameworks, the words, a delivery note.
"Drill me on BLUF."
→ practice: a scenario with hard numbers, then silence until you answer.
The dividing line is whether you want to attempt something or receive something.
Codex also lets you invoke one explicitly with $executive-presence or $executive-presence-practice when it doesn't pick up on its own; Claude Code triggers on the description alone. More in docs/USAGE.md, with worked transcripts in examples/.
What's in here
install.sh Installs for Claude Code, Codex, or both
skills/ The two skills — the installable artifact, read by both agents
executive-presence/
SKILL.md
references/frameworks.md the 12 frameworks
evals/evals.json test cases used to benchmark the skill
executive-presence-practice/
SKILL.md
references/frameworks.md identical copy — skills must be self-contained
.claude-plugin/ Claude Code plugin packaging (Codex ignores this)
prompts/ Portable versions for chat interfaces with no skill support
examples/ Worked transcripts showing what good output looks like
benchmarks/ Measured with-skill vs without-skill results
handouts/ Printable PDF of the frameworks, plus its build script
docs/ Install and usage guides
index.html The GitHub Pages landing page
robots.txt }
sitemap.xml } crawler files for the Pages site
.nojekyll }
On index.html
The landing page restates the 12 frameworks in HTML so the content is indexable — search engines don't read .md files on github.com as well as they read a served page. Nothing generates it, so if you change the frameworks, change the page too. The Markdown in skills/ stays canonical; it's what the skills actually read.
On the duplicated frameworks.md
Both copies are byte-identical and must stay that way — a skill has to be self-contained so it still works when someone copies one folder out. Verify with:
diff skills/executive-presence/references/frameworks.md skills/executive-presence-practice/references/frameworks.md
Does it actually work?
Measured against held-out scenarios, with and without the skill loaded. Full numbers and caveats in benchmarks/.
| Skill | Pass rate with skill | Without | Delta |
|---|---|---|---|
executive-presence | 100% ± 0% | 47% ± 46% | +53pp |
executive-presence-practice | 100% ± 0% | 87% ± 23% | +13pp |
The coach shows the larger gain, and the ±46% spread on the baseline is the interesting part: unaided, the model handled these scenarios inconsistently rather than uniformly badly. The practice skill starts from a higher baseline because "give me a scenario" is an easier instruction to follow unaided — its value is in the one rule a model breaks constantly without it: don't show the answer first.
Three evals per skill, three runs each. That's enough to show direction, not enough for a confident effect size.
License
MIT — see LICENSE.
The frameworks themselves are drawn from published work in communication and decision-making: Barbara Minto's Pyramid Principle, Edward de Bono's PMI, BLUF from military briefing practice, and SCR from consulting practice. The synthesis, examples, and skill instructions here are original.