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 swan-gtm/gtm-skills --skill human-mannerismsgit clone --depth 1 https://github.com/swan-gtm/gtm-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/swan-gtm/gtm-skills/human-mannerisms)<a href="https://agentmods.dev/skills/swan-gtm/gtm-skills/human-mannerisms"><img src="https://agentmods.dev/badge/skills/swan-gtm/gtm-skills/human-mannerisms/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/swan-gtm/gtm-skills/human-mannerisms"><img src="https://agentmods.dev/badge/skills/swan-gtm/gtm-skills/human-mannerisms.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 | $0.00131 | $0.01461 |
| Opus 5 | $0.00066 | $0.00731 |
| Sonnet 5 | $0.00026 | $0.00292 |
| Haiku 4.5 | $0.00013 | $0.00146 |
Grade A, and why
human-mannerisms 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 9d 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 — 55 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Applies when a piece of copy has had its AI tells removed but still has no voice. Produces a version with one or two earned human moves added, so a real person is visibly in the room.
Stripping AI tells gets copy "to zero": nothing screams machine. It does not get you above zero. Clean copy with no texture still reads as AI, because "flawless and flat" is itself the tell. This skill is the positive, additive pass: a small library of moves that put a specific person back into the writing. It runs after the subtractive pass, never instead of it.
This is deliberately a separate skill from the anti-ai-slop-writing pass, not one combined step, because the two jobs do not always both apply. Every piece that ships should have its AI tells removed; not every piece should be made to sound casual. A formal announcement, a compliance note, or a technical one-pager needs to be clean, not chatty - it gets the tell-stripping pass and stops there. A LinkedIn post, a founder's thought leadership, or a personal email needs voice on top - it gets both, in order. Bundling the two would force human texture onto writing that should stay plain. Keeping them apart lets you run the floor everywhere and reach for these moves only where the register calls for it.
The play
-
Strip first, then add. Run the de-AI pass before this one. Order matters: adding texture to copy that still has rhetorical AI tells just makes the tells louder. If the tells are gone, continue.
-
Read the piece once and locate the flatness. Find the stretch where the writing is smooth, correct, and voiceless - usually the middle, where the argument goes on autopilot. That is where a move earns its place. Do not spray moves across every line.
-
Add one or two moves, no more. This counts distinct move-types from the library below - pick one or two, not all eight. A single move-type may recur two or three times within the piece and still count as one move (the conjunction opener, for instance). The one exception is the mid-sentence aside: use it at most once per piece. What reads as try-hard is stacking many different quirks, so hold the line at one or two move-types per piece. Match the move to the writer's real register - a dry operator gets a deadpan undercut, not a whimsical aside.
-
Pull the real detail, never invent it. The specificity move (below) uses only true detail the writer actually has. Specificity you own reads as lived; specificity you invent reads as fraud, and invented numbers or moments are a hard failure.
The move library, with why each one works and how to place it, lives in references/move-library.md. The rules for pairing this pass with tell-stripping - the order, the "to zero / above zero" model, and what not to do - are in references/pairing-and-order.md.
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.
- 9d ago First seen · 55 lines · 131 tokens per session scan A 6b2c5eef4105
human-mannerisms is a skill published in the GitHub repository swan-gtm/gtm-skills (150 stars, last pushed 2d ago), licensed MIT. It adds 131 tokens to every session and 1,461 once invoked, about $0.0007 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 2026-09-03.
Other skills, from other repositories
four-ps-marketing
Designs and audits marketing strategy using McCarthy and Kotler's Four Ps framework: Product, Price, Place, and Promotion. Evaluates coherence across all four marketing mix elements to identify misalignments that undermine go-to-market effectiveness. Use when launching a new product, diagnosing why conversion rates…
Marketing Strategist
Marketing Strategist delivers practical business output with market context, metrics, stakeholder clarity, and execution focus.
lorekeeper-marketing
Marketing workflow for Lorekeeper — README copy, manifesto, positioning, and multi-agent A/B testing of copy variations. Load this when working on any user-facing content: README, docs/manifesto.md, positioning, comparison tables, or launch copy.
copy
Professional copywriter for SaaS and startups. Expert in landing page copy, positioning, messaging, conversion optimization, and voice-of-customer research. Use when writing compelling copy for SaaS products, landing pages, marketing materials, or when you need help with product positioning and messaging strategy.
marketing-strategy
Marketing strategy, positioning, personas, and channel planning with mandatory discovery gates.
magazine-web-ppt
For marketing and gtm work: bind launches, campaigns, events, and brand plans to growth and pipeline outcomes. Built around the core query "annual-marketing-plan", with GTM strategy lead judgment, buyer-ready proof, and this outcome: approve launch plan, campaign budget, or GTM motion.