Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Apply classical rhetoric — Ethos, Pathos, Logos — to analyze persuasive communication and craft effective arguments. Use this skill when the user needs to make a speech more persuasive, analyze why a piece of communication is effective, write a compelling proposal, or evaluate rhetorical strategies — even if they say 'make this more convincing', 'why is this speech so powerful', or 'how do I persuade the board'.
.claude/skills/asgard-ai-platform-hum-rhetoric/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -5% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 3% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 17% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 3% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 56% | 0% |
Aristotle identified three modes of persuasion: Ethos (credibility), Pathos (emotion), and Logos (logic). Effective persuasion uses all three in proportion appropriate to the audience and context.
IRON LAW: All Three Appeals, Calibrated to Audience
Logos alone convinces analysts but bores executives. Pathos alone moves
hearts but lacks substance. Ethos alone relies on reputation that may
not exist.
Every persuasive communication must use all three, weighted by audience:
- Technical audience: Lead with Logos, support with Ethos
- Executive audience: Lead with Pathos (vision), support with Logos (data)
- Public audience: Lead with Ethos (trust), amplify with PathosEthos (Credibility) — Why should they trust YOU?
Pathos (Emotion) — Why should they CARE?
Logos (Logic) — Why should they BELIEVE?
| Device | What It Does | Example | |--------|-------------|---------| | Anaphora | Repeating the opening phrase | "We will fight... We will never surrender... We will..." | | Tricolon | Group of three | "Life, liberty, and the pursuit of happiness" | | Antithesis | Juxtaposing opposites | "Ask not what your country can do for you, ask what you can do for your country" | | Rhetorical question | Question with an obvious answer | "Can we really afford to wait?" | | Metaphor/Analogy | Comparing abstract to concrete | "This project is our moonshot" | | Chiasmus | Reversed parallel structure | "We don't stop when we're tired; we stop when we're done" |
markdown# Rhetorical Analysis: {Text/Speech} ## Rhetorical Situation - Speaker: ... - Audience: ... - Purpose: ... - Context: ... ## Appeal Analysis | Section | Appeal | Technique | Effectiveness | |---------|--------|-----------|--------------| | {quote/section} | Ethos/Pathos/Logos | {specific technique} | H/M/L | ## Overall Balance - Ethos: {strong/weak} — {evidence} - Pathos: {strong/weak} — {evidence} - Logos: {strong/weak} — {evidence} ## Verdict {Overall effectiveness and recommendations for improvement}
Scenario: Analyzing a startup pitch
references/speech-structures.md| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 25,618 | 26,894 | +5% | 1 | 1 | 0% | 3,956 | 3,769 | -5% | 0 | 0 | — |
case-02 | fail→pass | 47,001 | 18,153 | -61% | 1 | 1 | 0% | 4,500 | 4,646 | +3% | 0 | 0 | — |
case-03 | fail→fail | 32,334 | 17,555 | -46% | 1 | 1 | 0% | 3,252 | 4,147 | +28% | 0 | 0 | — |
case-04 | fail→pass | 16,484 | 14,374 | -13% | 1 | 1 | 0% | 2,752 | 3,220 | +17% | 0 | 0 | — |
case-05 | fail→pass | 16,129 | 10,558 | -35% | 1 | 1 | 0% | 2,810 | 2,883 | +3% | 0 | 0 | — |
case-06 | fail→pass | 13,891 | 16,022 | +15% | 1 | 1 | 0% | 2,426 | 3,791 | +56% | 0 | 0 | — |
case-07 | pass→pass | 9,034 | 9,471 | +5% | 1 | 1 | 0% | 1,524 | 2,617 | +72% | 0 | 0 | — |
case-08 | pass→pass | 13,877 | 13,523 | -3% | 1 | 1 | 0% | 2,264 | 3,646 | +61% | 0 | 0 | — |
case-09 | pass→pass | 9,135 | 9,902 | +8% | 1 | 1 | 0% | 1,462 | 2,649 | +81% | 0 | 0 | — |
case-10 | fail→pass | 19,630 | 11,665 | -41% | 1 | 1 | 0% | 2,941 | 3,050 | +4% | 0 | 0 | — |
case-11 | pass→pass | 13,225 | 10,273 | -22% | 1 | 1 | 0% | 1,871 | 2,694 | +44% | 0 | 0 | — |
case-12 | fail→pass | 12,403 | 10,896 | -12% | 1 | 1 | 0% | 1,966 | 2,811 | +43% | 0 | 0 | — |
case-13 | pass→pass | 20,741 | 20,418 | -2% | 1 | 1 | 0% | 2,935 | 3,848 | +31% | 0 | 0 | — |
case-14 | fail→pass | 17,267 | 15,343 | -11% | 1 | 1 | 0% | 2,573 | 3,667 | +43% | 0 | 0 | — |
case-15 | fail→pass | 29,996 | 15,402 | -49% | 1 | 1 | 0% | 1,915 | 3,363 | +76% | 0 | 0 | — |
case-16 | fail→fail | 16,711 | 12,800 | -23% | 1 | 1 | 0% | 2,408 | 3,273 | +36% | 0 | 0 | — |
case-17 | fail→pass | 26,329 | 13,795 | -48% | 1 | 1 | 0% | 3,823 | 3,149 | -18% | 0 | 0 | — |
case-18 | pass→pass | 14,514 | 10,119 | -30% | 1 | 1 | 0% | 2,043 | 2,812 | +38% | 0 | 0 | — |
case-19 | pass→pass | 19,019 | 15,978 | -16% | 1 | 1 | 0% | 3,088 | 3,879 | +26% | 0 | 0 | — |
case-20 | pass→pass | 5,613 | 3,369 | -40% | 1 | 1 | 0% | 800 | 1,628 | +103% | 0 | 0 | — |
case-21 | fail→fail | 15,059 | 13,774 | -9% | 1 | 1 | 0% | 2,172 | 3,210 | +48% | 0 | 0 | — |
case-22 | pass→pass | 15,852 | 15,697 | -1% | 1 | 1 | 0% | 2,230 | 3,882 | +74% | 0 | 0 | — |
case-23 | pass→pass | 7,081 | 4,366 | -38% | 1 | 1 | 0% | 1,007 | 1,922 | +91% | 0 | 0 | — |
case-24 | pass→fail | 9,437 | 12,134 | +29% | 1 | 1 | 0% | 1,848 | 3,419 | +85% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 24 cases were attempted. The headline lift of +38 percentage points is the difference between those two pass rates over the 24 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.