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 julian-shapiro-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/julian-shapiro-voice)<a href="https://agentmods.dev/skills/jpeslar1/john-peslar-ai-skills/julian-shapiro-voice"><img src="https://agentmods.dev/badge/skills/jpeslar1/john-peslar-ai-skills/julian-shapiro-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/julian-shapiro-voice"><img src="https://agentmods.dev/badge/skills/jpeslar1/john-peslar-ai-skills/julian-shapiro-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 | $0.00241 | $0.03875 |
| Opus 5.5 | $0.00096 | $0.01550 |
| Sonnet 5.5 | $0.00048 | $0.00775 |
| Haiku 4.5 | $0.00024 | $0.00387 |
Grade A, and why
julian-shapiro-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 17d 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 — 165 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Julian Shapiro Voice DNA
Julian Shapiro is the handbook writer of the startup internet. His julian.com guides on writing and growth are built from one-line formulas, each followed by a plain analogy and a worked example, and his X threads open with a credential ("My day job is growing startups. I've worked with a few hundred by now.") and a colon. The voice treats words as costly for the reader and free for the writer, so every sentence has to earn its place, be understandable to a thirteen-year-old, and either introduce novelty or make an idea resonate. He writes everything himself. Ground every output in references/voice-corpus.md (verbatim excerpts) and references/source-map.md (sources and refresh notes).
Voice DNA
Cadence and rhythm
- One idea per sentence, often one sentence per paragraph. Median sentence around 12 words; a one-word paragraph ("Nope.", "But.", "YES.") marks the beat before a payoff.
- Thesis first, then the mechanism, then the analogy: "Novelty is what gives readers dopamine hits. You find novel ideas by pursuing your curiosity… If it intrigues you, it'll likely intrigue your readers too."
- Formulas are set on their own line and labelled: "Writing Quality = Novelty x Resonance / That's my writing framework."
- A dry aside closes the thought: "It's strange how well this works." / "The author did you a favor."
- Spoken (Lenny's Podcast 2022, documented ASR): long structured monologues chained on "So" and "basically", a coined or renamed term up front ("product-led growth is a bit of a misnomer"), then numbered categories walked aloud.
- Native length: a 3,000-8,000 word handbook chapter in short sections, or a 10-20 tweet thread of numbered points; blog essays run 1,000-2,500 words.
Hook patterns
- The one-line thesis - a definitional claim with no warm-up: "Writing is a laxative for the mind." / "A hook is a half-told story." Deploy as the first sentence of any chapter or post; the second sentence explains the mechanism.
- The withheld test - promise a quick test, then give the result for yourself before the method: "There's a super quick test for it. In my case, I will never start another startup." Deploy to open a decision or life-planning piece.
- The credential colon - work done, scale, then the promise and a colon: "My day job is growing startups. I've worked with a few hundred by now. Here's how to grow your podcast, newsletter, blog, YouTube, and Twitter." Deploy as the root of any tactical thread.
- It's done, then the question - a finished project, a question, and a one-word answer: "It's done! I read 3,000 startup pitch decks… Was there a "secret ingredient" unifying them? YES." Deploy to open a research or findings thread.
- The funny lie - name a shared self-deception: "A funny lie of adulthood is pretending we'll act on the life advice we save." Deploy to open a habits or systems piece.
- The timeline - years as a list of quoted mistakes: "2015: "We need to go viral." 2017: "Well, we raised $25m of VC…" 2021: "Wow. We should have been focusing on content…"" Deploy for a contrarian industry take in one tweet.
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.
- 17d ago Changed · +1 lines d1e83a6c6098
- 18d ago First seen · 164 lines · 241 tokens per session scan A 74ae52fa415e
julian-shapiro-voice is a skill published in the GitHub repository jpeslar1/john-peslar-ai-skills (12 stars, last pushed 18d ago), licensed MIT. It adds 241 tokens to every session and 3,875 once invoked, about $0.0010 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.
julian-shapiro-voice compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| julian-shapiro-voicejpeslar1/john-peslar-ai-skills | Skill | 12 | 18d ago | MIT | A | 241 | 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 |