Install a Department, Not a Prompt, with headcount

One prompt cannot carry a company's worth of domain knowledge. headcount — a 2k-star MIT repository by Chris Brock — structures it as a company instead: a chief executive over 16 departments, 172 skills in total, each department an independently installable plugin, so a project loads only the functions it needs rather than all of them at once.

Why This Skill Matters

The departments cover the whole org chart, and the README's own department roster announces the counts:

Two structural ideas separate this from a flat skill list. First, skills are addressed as department:skill — security:threat-modeling, finance:unit-economics — so names never collide. Second, the security and legal-risk departments are reviewer-class: they review what other departments build, and their blocking findings are not overrulable by the department under review. That is why the CISO and the CLO report to the chief executive rather than into the function they oversee.

Skills also stay grounded. Those answering questions an outside authority settles carry the authorities in references/sources.md inside the skill — 184 cited sources across 150 skills, spanning tax and accounting, law and employment, security and controls frameworks, and education standards. The README's rule is "references, never copies": 125 of the 184 are quotable, the rest are read-and-cite, and each entry says which in the imperative next to the link.

Installation

In Claude Code, add the marketplace and install one department:

/plugin marketplace add cbrock84/headcount
/plugin install security@headcount

Install as many departments as the project needs. ChatGPT and Codex read the same repository — add it as a plugin marketplace, or drop the department you want into .agents/skills/ in your own project. The skills are identical in both; only the manifests differ, and both sets are generated from the same tree, so a fix reaches both at once.

Real Workflow: A Threat Model Before You Build

1. Install the security department with the two commands above.
2. Ask a question in the department's territory — for example, "review this design before we build it". The skill loads itself: the README's matching-table example routes that request to security:threat-modeling.
3. threat-modeling identifies what could go wrong before anything is built or changed: the assets worth attacking, the entry points, the trust boundaries, and the controls that address them.
4. When the fix spans departments — the README's own cross-department example is an enterprise prospect demanding SOC 2 — the worked end-to-end situations live in docs/USE-CASES.md, including where a reviewer-class department stops the work rather than adding an opinion.
5. Delegate a whole department as a subagent: each department ships an agent charter in .claude/agents/ with its own exclusive write surface.

If you would rather pick a specific lens than wait for auto-loading, invoke it directly: /finance:financial-modeling.

Tips

• Skills load on match, so install only what the project needs — the README's rule is to install as many departments as the project needs, and docs/GETTING-STARTED.md covers which departments to install first, the three ways to invoke a skill, and what reviewer-class departments do differently.
• Split by write surface, not by topic — the README's stated principle is that agents split by exclusive write surface, because a topic split has no checkable boundary and two agents working on "SEO" and "UI" both end up in the same file. The method itself is documented in executive:agent-hierarchy.
• Use the interactive org chart at cbrock84.github.io/headcount/org-chart.html to search every skill and jump to its source before deciding which departments to install.
• Everything is checkable locally: ./scripts/check-all.sh runs every check CI runs — frontmatter validity, department:skill reference resolution, and manifest parity between Claude and ChatGPT/Codex among them — so local and CI cannot drift.

When Not to Use This

The repo's own design rule cuts against installing everything: departments exist so that a project loads only the functions it needs, rather than all of them at once. And where a skill stops at advice and the work needs a system, the README lists companion products from the same builder — a GRC platform for SOC 2, ISO 27001, NIST CSF, and HIPAA programs, an AI-answer scanner, and an app builder — while stating plainly that nothing in the repository requires them.

See the leaderboard for more skills.