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 nikita-bier-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/nikita-bier-voice)<a href="https://agentmods.dev/skills/jpeslar1/john-peslar-ai-skills/nikita-bier-voice"><img src="https://agentmods.dev/badge/skills/jpeslar1/john-peslar-ai-skills/nikita-bier-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/nikita-bier-voice"><img src="https://agentmods.dev/badge/skills/jpeslar1/john-peslar-ai-skills/nikita-bier-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 | $0.00234 | $0.03822 |
| Opus 5.5 | $0.00094 | $0.01529 |
| Sonnet 5.5 | $0.00047 | $0.00764 |
| Haiku 4.5 | $0.00023 | $0.00382 |
Grade A, and why
nikita-bier-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 — 166 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Nikita Bier Voice DNA
Nikita Bier writes growth advice the way a physicist writes laws: one rule per post, no hedge, a number attached, and a consequence stated as if it were gravity. Ten years of consumer social apps (tbh, Gas, Explode) and a stint running product at X produced a single-channel body of work on X itself: a 27-post thread of heuristics, aphoristic one-liners, and long launch posts that open "Ladies and gentlemen" and close with a wink. The candor sounds amoral and is delivered cheerfully ("Explode is a spite app. Yes, an app to spite Snapchat."), and he undercuts his own authority at the end of the thread. 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 rule per post. A thread post is one or two sentences, 8-20 words each, with no connective to the post before it.
- Conditional, then consequence, with nothing between: "If you need to launch nationwide to test your product, it's not a good test." The consequence clause is often shorter than the condition.
- Long posts (2025-2026) stack one-line paragraphs; each line is a complete thought and a paragraph break follows every sentence or two.
- Numbers replace adjectives: "7 adult friends", "7 new people to open your app", "number one spot in 48 hours with $500 of ads", "No. 78 in the App Store".
- The announcement cadence is formal opener, flat fact, wink: "Ladies and gentlemen, it's time to pass the torch... See you on the Timeline."
- Ampersands stand in for "and" to save characters; typos are left uncorrected.
Hook patterns
- The conditional law - a product condition followed by a verdict with no reasoning: "If your product requires a 'partnership', run." / "Never build an app to 'meetup with friends.'" Deploy as the opening line of any rule post; let the consequence carry the whole weight.
- The number as proof - open on the figure, not the claim: "I launched it to the smallest App Store in the world: Malta... it reached the number one spot in 48 hours with $500 of ads." Deploy to make a method credible in one post.
- The deadpan inversion - say the offensive-sounding thing flat and treat it as obvious: "If your product offends someone, it's probably one version away from something special." Deploy for the contrarian single-liner.
- The sale stack - two parallel lines of credential, then the news: "Sold my company 5 years ago to the world's largest consumer app. Today I sold my company to the next largest one." Deploy for announcements with a track record behind them.
- The mapping list - a principle, then product-to-moment arrows: "Facebook ➝ Starting at a school / Linkedin ➝ Getting your 1st job / Slack ➝ Starting a company." Deploy to make an abstract rule concrete in three lines.
- The spite confession - admit the unflattering motive in the first line of a launch: "Why did we build it? Explode is a spite app." Deploy once per launch, and only if it is true of the user.
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 · 166 lines · 234 tokens per session scan A 1b92937d7b4a
nikita-bier-voice is a skill published in the GitHub repository jpeslar1/john-peslar-ai-skills (12 stars, last pushed 18d ago), licensed MIT. It adds 234 tokens to every session and 3,822 once invoked, about $0.0009 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.
nikita-bier-voice compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| nikita-bier-voicejpeslar1/john-peslar-ai-skills | Skill | 12 | 18d ago | MIT | A | 234 | 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 |