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 deciqAI/knowledge-skills --skill dunbars-numbergit clone --depth 1 https://github.com/deciqAI/knowledge-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/deciqai/knowledge-skills/dunbars-number)<a href="https://agentmods.dev/skills/deciqai/knowledge-skills/dunbars-number"><img src="https://agentmods.dev/badge/skills/deciqai/knowledge-skills/dunbars-number/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/deciqai/knowledge-skills/dunbars-number"><img src="https://agentmods.dev/badge/skills/deciqai/knowledge-skills/dunbars-number.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 106 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00115 | $0.01909 |
| Opus 5 | $0.00057 | $0.00955 |
| Sonnet 5 | $0.00023 | $0.00382 |
| Haiku 4.5 | $0.00012 | $0.00191 |
Grade A, and why
dunbars-number 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 8d 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 — 124 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Dunbar's Number
Overview
Dunbar's number (~150) is the cognitive limit on stable social groups the human neocortex can sustain, with nested layers at ~5, ~15, ~50, then outer bands at ~500 and ~1,500. Derived by Robin Dunbar (1992) from primate neocortex ratios. Key implication: below ~150, informal trust coordinates behavior; above ~150, formal hierarchy becomes architecturally mandatory — not a cultural choice.
Composes with network-effects (different coordination mechanisms below vs. above 150), principal-agent (exceeding 150 multiplies agency problems informal trust can no longer cover), founder-mindset (founder relationship bandwidth peaks at ~150 — first breakpoint demanding shift from doing to delegating structure).
When to Use
- Headcount approaching or crossing 150 (or 50, or 500)
- "Coordination getting harder" complaints with no obvious structural cause
- Deciding to split a team, open a new office, or create a sub-unit
- Designing org structure for a scaling startup; evaluating acquisition integration plans
- Diagnosing why informal culture "felt different" after a headcount milestone
- AI-native startup hiring fast on abundant AI-era capital / AI-adoption demand — "we scaled from 20 to 150 in a year and coordination is breaking"
Not when: group well below 50 and problems are interpersonal; constraint is incentive misalignment (use principal-agent); challenge is growing a network beyond 150 (use network-effects).
Coaching Novices (Adaptive Front Door)
- Engine mode: user has a concrete org design or headcount question → run The Process directly.
- Coach mode: user is unfamiliar → guide step by step.
In Coach mode, respond one step at a time. Each [WAIT] is a hard stop — output only that step's question, then stop.
- One-liner: once a group exceeds ~150 people, informal trust cannot coordinate it — formal structure is architecturally required by the human neocortex.
- Check fit: what is the current headcount? Which Dunbar layer, which breakpoint is approaching?
- Elicit the structure: what coordination mechanisms exist? Where are they breaking down?
[WAIT — do not advance until user responds]
- Run The Process one step at a time with their input.
[WAIT — do not advance until user responds]
- Close: group layer + breakpoint diagnosis + structural prescription (split / formalize / layer) named.
[WAIT — do not advance until user responds]
What ships with it
3 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.
- 8d ago First seen · 124 lines · 115 tokens per session scan A bbee891df2b1
dunbars-number is a skill published in the GitHub repository deciqAI/knowledge-skills (10 stars, last pushed 10d ago), licensed MIT. It adds 115 tokens to every session and 1,909 once invoked, about $0.0006 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
cro
Conversion audit for funnels and pages — a section-by-section friction log (clarity, anxiety, distraction, motivation), heuristic checks (message match, above-the-fold value prop, form cost), an ICE-scored hypothesis backlog, and top-3 A/B test designs with success metrics and minimum-sample notes. Use when the user…
prompt-master
Generates optimized prompts for AI tools. Activates only when the user explicitly asks to write, fix, improve, or adapt a prompt for a specific AI tool (LLM, Cursor, Midjourney, image AI, video AI, coding agents, etc.). Does not activate for general conversation, coding tasks, document writing, or other…
slack-tools
Slack workspace management and automation specialist.
syndic
Gère un parc de copropriétés en France avec vue portfolio consolidée. Couvre administration, comptabilité (décret 2005, plan comptable copro, 5 annexes), assemblées générales (convocation, PV, notification), appels de fonds, travaux, fournisseurs, recouvrement d'impayés et transition de syndic. Maîtrise les majorités…
ops-suggest
Show time-aware CocoOps operational suggestions from the deterministic ops-suggest classifier. Usage: $ops suggest.
customer-onboarding-and-implementation
Takes a new customer from signature to working — setting a definition of live that both sides agreed before the contract was signed, planning and staffing the implementation, running data migration and integration realistically, training the people who will actually use it, and handing over to the ongoing…