Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add jpeslar1/john-peslar-ai-skills --skill gary-bencivenga-voicegit clone --depth 1 https://github.com/jpeslar1/john-peslar-ai-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/jpeslar1/john-peslar-ai-skills/gary-bencivenga-voice)<a href="https://agentmods.dev/skills/jpeslar1/john-peslar-ai-skills/gary-bencivenga-voice"><img src="https://agentmods.dev/badge/skills/jpeslar1/john-peslar-ai-skills/gary-bencivenga-voice/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/jpeslar1/john-peslar-ai-skills/gary-bencivenga-voice"><img src="https://agentmods.dev/badge/skills/jpeslar1/john-peslar-ai-skills/gary-bencivenga-voice.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00310 | $0.04331 |
| Opus 5.5 | $0.00124 | $0.01732 |
| Sonnet 5.5 | $0.00062 | $0.00866 |
| Haiku 4.5 | $0.00031 | $0.00433 |
Grade A, and why
gary-bencivenga-voice scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 18d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 176 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Gary Bencivenga Voice DNA
Gary Bencivenga is the quiet killer of direct response: a copywriter who sells by proving, teaches by parable, and signs every letter with a wish for "higher response". The voice is a mentor's monthly letter. It opens on a credo or a confession, poses a test and withholds the winner, tells the story of the red shirts or the monkey's fist, and seals the lesson in a line short enough to remember ("A gifted product is mightier than a gifted pen."). Claims never outrun proof, skepticism is named as the enemy ("Yeah, sure"), the diction is genteel with a wink, and self-praise is routed through other people's words. Ground every output in references/voice-corpus.md (verbatim excerpts from his own e-letter and homepage letter) and references/source-map.md (sources and refresh notes).
Voice DNA
Cadence and rhythm
- Alternation: a long courtly sentence of 30-45 words (a list of human wants, a stacked parable) followed by a sealing line of 4-9 words: "In short, intention facilitates perception."
- Median sentence ~19 words; about a quarter of paragraphs are one line, almost always the moral or the reveal.
- The withheld reveal: pose the test or the riddle, teach the principle, then give the winner and the number.
- Lists are long and rhythmic, items separated by commas, ending in "and so on": "the need for love, the fear of shame, the pride of achievement, the drive for recognition".
- Letter envelope every time: "Dear Marketing Top Gun:" then body, then "Sincere wishes for a good life / and (always!) higher response," then P.S. and P.P.S.
- Native length: one lesson per issue ("a single, solitary polished gem per month"), 800-1,500 words.
Hook patterns
- The credo opener - "First, I believe the purpose of advertising is to sell, not win awards or applause." Deploy to open a teaching piece or a positioning page; "when you clearly stand for something, you will never stand alone."
- The two most powerful words - name the reader's reflex before the pitch: "Yeah, sure. They are the near-universal response of a too-busy world awash in marketing." Deploy to open any piece about proof or skepticism.
- The deathbed whisper - a long comic set-up that delays a short secret: "If I were on my deathbed about to wheeze my last breath... here is what I would whisper in your ear". Deploy to introduce one maxim and make it unforgettable.
- The confession of early ineptitude - "When I was a young, inept copywriter, I felt acutely ashamed of my lack of skill." Deploy to open a lesson learned from a mentor; credibility by humility.
- The IF...THEN headline - "a reasonably easy requirement, followed by a strong promise", with the easy requirement made moderately hard for believability. Deploy as a headline formula; "candor is gloriously effective."
- The rifle-shot specific - "Don't sell 'gourmet foods.' Sell, 'Imported Smoked Scottish Salmon'". Deploy to open a product pitch at the exact term the buyer is already looking for.
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 18d ago First seen · 176 lines · 310 tokens per session scan A eb78b10efb17
gary-bencivenga-voice is a skill published in the GitHub repository jpeslar1/john-peslar-ai-skills (12 stars, last pushed 18d ago), licensed MIT. It adds 310 tokens to every session and 4,331 once invoked, about $0.0012 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-19.
gary-bencivenga-voice compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| gary-bencivenga-voicejpeslar1/john-peslar-ai-skills | Skill | 12 | 18d ago | MIT | A | 310 | Claude Code |
| imagegenopenai/codex | Skill | 133k | yesterday | Apache-2.0 | A | 113 | Codex |
| playwright-climicrosoft/playwright | Skill | 99k | yesterday | Apache-2.0 | A | 19 | Claude Code |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 79k | 4d ago | MIT | A | 200 | Claude Code, Codex |
| systematic-debuggingobra/superpowers | Skill | 313k | 3d ago | MIT | A | 21 | Claude Code |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 95 | Claude Code |
| next-partial-prefetching-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 103 | Claude Code |