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 naveedharri/benai-skills --skill bens-voicegit clone --depth 1 https://github.com/naveedharri/benai-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/naveedharri/benai-skills/bens-voice)<a href="https://agentmods.dev/skills/naveedharri/benai-skills/bens-voice"><img src="https://agentmods.dev/badge/skills/naveedharri/benai-skills/bens-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/naveedharri/benai-skills/bens-voice"><img src="https://agentmods.dev/badge/skills/naveedharri/benai-skills/bens-voice.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00109 | $0.02022 |
| Opus 5 | $0.00055 | $0.01011 |
| Sonnet 5 | $0.00022 | $0.00404 |
| Haiku 4.5 | $0.00011 | $0.00202 |
Grade A, and why
bens-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 7d 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 — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.
bens-voice: the Ben Voice Engine
Two jobs: draft content as Ben, and judge any draft against his voice. Never ship a draft that has not been through the judge loop below.
Built from his real corpus (May-July 2026): 10 YouTube transcripts (~47,700 spoken words), 88 posted Circle comments, 10 LinkedIn posts, 6 newsletter issues, 78 Slack messages, and 45 LinkedIn DMs. The references are the measured result; trust them over instinct. When instinct disagrees with a number in a reference file, the number wins.
Workflow
- Classify the content type:
circle-reply|linkedin-post|newsletter|slack|dm|email|youtube-script|generic. - Read the references for that type (table below). Non-negotiable.
- Circle replies only: classify the POST type first via the reply-type playbook inside
references/voice-circle.md(it decides length and shape), and use its Routing section for Milan/Q&A/course conventions. - Draft.
- Lint (Layer 1, deterministic). Run the bundled script from this skill's directory (paths below are relative to the folder containing this SKILL.md):
python3 scripts/voice_lint.py --type <type> <<'EOF' <draft text> EOF - Judge (Layer 2): score the draft with the rubric below, honestly, dimension by dimension. Any HARD lint violation caps the total at 59 regardless of rubric math.
- Score below 80: fix the named failures and redraft. Maximum 2 redrafts, then output the best attempt with its score.
- Output the final draft plus this verdict block:
--- BEN VOICE ENGINE VERDICT --- score: NN/100 (directness N/20, register N/20, opinions N/20, specificity N/15, structure N/15, length N/10) lint: PASS|FAIL (N hard, N soft: rule names) attempts: N
Reference map
| Type | Read |
|---|---|
| circle-reply | references/voice-circle.md + references/gold-circle.md + references/values.md |
| linkedin-post | references/voice-linkedin.md + references/values.md |
| newsletter | references/voice-newsletter.md + references/values.md |
| slack | references/voice-slack.md + references/values.md |
| dm, email | references/voice-dm.md + references/values.md |
| youtube-script | references/voice-youtube.md + references/values.md |
| generic | closest match above + references/values.md |
What ships with it
11 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.
- CALIBRATION.md 2.8 KB
- references/gold-circle.md 7.9 KB
- references/maintenance.md 2.1 KB
- references/values.md 7.3 KB
- references/voice-circle.md 19 KB
- references/voice-dm.md 4.1 KB
- references/voice-linkedin.md 22 KB
- references/voice-newsletter.md 21 KB
- references/voice-slack.md 11 KB
- references/voice-youtube.md 24 KB
- scripts/voice_lint.py 13 KB runs code
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.
- 7d ago First seen · 89 lines · 109 tokens per session scan A d52b75e3a9d4
bens-voice is a skill published in the GitHub repository naveedharri/benai-skills (61 stars, last pushed yesterday), licensed MIT. It adds 109 tokens to every session and 2,022 once invoked, about $0.0005 per session on Opus 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 2026-09-05.
Other skills, from other repositories
project
A single starting point for setting up an AI-assisted project in Claude Cowork. It asks questions about the work, reviews available add-ons, and creates project instructions, custom agents, and connected workflows.
design-sync-upload
An uploader for design-system files such as DESIGN.md, tokens, logos, fonts, and images into Claude Design. It can either use an authenticated connection or prepare a folder and guide for manual upload.
doc-html-slide
A renderer that turns presentation content into a single HTML slide deck that opens directly in a browser. It creates a 16:9 slide sequence with navigation, fullscreen viewing, printing to PDF, and speaker-note controls.
cs-channel-message
A channel-message writing tool for search ads, advertising, customer relationship messages, and app notifications. It uses the NCM sequence—Need, Channel, Moment, Message, and CTA—to adapt wording to where and when customers see it.
design-tokens-transformer
A converter for design tokens, which are named values for colors, fonts, spacing, borders, shadows, and motion. It translates one shared token source into CSS variables and Tailwind or shadcn-style theme files, and can convert them back for checking.
media-higgsfield-explainer
A Higgsfield workflow for making non-photorealistic narrated explainer videos. It pairs each narration line with a 10-second animated clip and joins the clips into one finished video.