Skip to content

An agent organization structured as a company — 15+ departments, 125+ skills, each independently installable, citing the standards and regulators that settle the question. Runs in Claude Code and ChatGPT.

Install in ChatGPT and Codex from the same tree

Claudec3eb59b401d8
diff --git a/.agents/plugins/marketplace.json b/.agents/plugins/marketplace.json
new file mode 100644
index 0000000..c8ae0b6
--- /dev/null
+++ b/.agents/plugins/marketplace.json
@@ -0,0 +1,184 @@
+{
+ "name": "headcount",
+ "interface": {
+ "displayName": "headcount"
+ },
+ "plugins": [
+ {
+ "name": "executive",
+ "source": {
+ "source": "local",
+ "path": "./plugins/executive"
+ },
+ "policy": {
+ "installation": "AVAILABLE"
+ },
+ "category": "Productivity"
+ },
+ {
+ "name": "technology",
+ "source": {
+ "source": "local",
+ "path": "./plugins/technology"
+ },
+ "policy": {
+ "installation": "AVAILABLE"
+ },
+ "category": "Productivity"
+ },
+ {
+ "name": "it-operations",
+ "source": {
+ "source": "local",
+ "path": "./plugins/it-operations"
+ },
+ "policy": {
+ "installation": "AVAILABLE"
+ },
+ "category": "Productivity"
+ },
+ {
+ "name": "product",
+ "source": {
+ "source": "local",
+ "path": "./plugins/product"
+ },
+ "policy": {
+ "installation": "AVAILABLE"
+ },
+ "category": "Productivity"
+ },
+ {
+ "name": "marketing",
+ "source": {
+ "source": "local",
+ "path": "./plugins/marketing"
+ },
+ "policy": {
+ "installation": "AVAILABLE"
+ },
+ "category": "Productivity"
+ },
+ {
+ "name": "demand-generation",
+ "source": {
+ "source": "local",
+ "path": "./plugins/demand-generation"
+ },
+ "policy": {
+ "installation": "AVAILABLE"
+ },
+ "category": "Productivity"
+ },
+ {
+ "name": "revenue",
+ "source": {
+ "source": "local",
+ "path": "./plugins/revenue"
+ },
+ "policy": {
+ "installation": "AVAILABLE"
+ },
+ "category": "Productivity"
+ },
+ {
+ "name": "finance",
+ "source": {
+ "source": "local",
+ "path": "./plugins/finance"
+ },
+ "policy": {
+ "installation": "AVAILABLE"
+ },
+ "category": "Productivity"
+ },
+ {
+ "name": "operations",
+ "source": {
+ "source": "local",
+ "path": "./plugins/operations"
+ },
+ "policy": {
+ "installation": "AVAILABLE"
+ },
+ "category": "Productivity"
+ },
+ {
+ "name": "pmo",
+ "source": {
+ "source": "local",
+ "path": "./plugins/pmo"
+ },
+ "policy": {
+ "installation": "AVAILABLE"
+ },
+ "category": "Productivity"
+ },
+ {
+ "name": "people",
+ "source": {
+ "source": "local",
+ "path": "./plugins/people"
+ },
+ "policy": {
+ "installation": "AVAILABLE"
+ },
+ "category": "Productivity"
+ },
+ {
+ "name": "legal-risk",
+ "source": {
+ "source": "local",
+ "path": "./plugins/legal-risk"
+ },
+ "policy": {
+ "installation": "AVAILABLE"
+ },
+ "category": "Productivity"
+ },
+ {
+ "name": "security",
+ "source": {
+ "source": "local",
+ "path": "./plugins/security"
+ },
+ "policy": {
+ "installation": "AVAILABLE"
+ },
+ "category": "Productivity"
+ },
+ {
+ "name": "customer-experience",
+ "source": {
+ "source": "local",
+ "path": "./plugins/customer-experience"
+ },
+ "policy": {
+ "installation": "AVAILABLE"
+ },
+ "category": "Productivity"
+ },
+ {
+ "name": "data-analytics",
+ "source": {
+ "source": "local",
+ "path": "./plugins/data-analytics"
+ },
+ "policy": {
+ "installation": "AVAILABLE"
+ },
+ "category": "Productivity"
+ },
+ {
+ "name": "corporate-strategy",
+ "source": {
+ "source": "local",
+ "path": "./plugins/corporate-strategy"
+ },
+ "policy": {
+ "installation": "AVAILABLE"
+ },
+ "category": "Productivity"
+ }
+ ]
+}
diff --git a/AGENTS.md b/AGENTS.md
new file mode 100644
index 0000000..aff0ecf
--- /dev/null
+++ b/AGENTS.md
@@ -0,0 +1,50 @@
+# AGENTS.md
+
+<!-- Generated by scripts/build-port.py. Do not edit. -->
+
+An agent organization structured as a company: 16 departments, 172 skills, each department installable on its own.
+
+This file describes the repository to an agent working *on* it. The skills under `plugins/` are a different thing — they describe how to do the work of a function, and they load when a request matches their description.
+
+## Installing, in either tool
+
+The skills are identical. Only the manifests differ, and both are generated from the same tree, so a skill fixed once is fixed for both.
+
+| | Manifests |
+|---|---|
+| Claude Code | `plugins/<department>/.claude-plugin/plugin.json`, `.claude-plugin/marketplace.json` |
+| ChatGPT / Codex | `plugins/<department>/.codex-plugin/plugin.json`, `.agents/plugins/marketplace.json` |
+
+## Before you open a pull request
+
+```
+./scripts/check-all.sh
+```
+
+Every check CI runs, in one script — the workflow calls this same file, so the two cannot drift. Stage new files first: the surface guard reads `git ls-files`, so an unstaged file is invisible to it and the check passes locally then fails in CI.
+
+## Rules that are enforced rather than suggested
+
+- **One owner per path.** `docs/AGENT-SURFACES.md` maps every tracked path to exactly one agent, and no two owners may claim the same path. A new department adds its roster row, its surface block and its charter in the same change.
+- **Generated files are never hand-edited.** The README, the org chart, the social card, each skill's `references/sources.md`, and these manifests are all emitted from the tree and verified with `--check`. An edit to one is lost on the next regenerate.
+- **US English**, by exact word form. `scripts/check-us-english.py --fix` rewrites.
+- **No third-party licensed content.** All prose here is original; `scripts/check-provenance.py` is the backstop.
+
+## Departments
+
+- `executive` — Office of the CEO
+- `technology` — Technology
+- `it-operations` — IT Operations
+- `product` — Product
+- `marketing` — Marketing
+- `demand-generation` — Demand Generation
+- `revenue` — Revenue
+- `finance` — Finance
+- `operations` — Operations
+- `pmo` — Program Management Office
+- `people` — People
+- `legal-risk` — Legal & Risk
+- `security` — Security
+- `customer-experience` — Customer Experience
+- `data-analytics` — Data & Analytics
+- `corporate-strategy` — Corporate Strategy
diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md
index 48bad41..b1b2f4f 100644
--- a/CONTRIBUTING.md
+++ b/CONTRIBUTING.md
@@ -61,6 +61,7 @@ Every check CI runs, in one script — the workflow calls this same file, so the
| Skill references | Every `department:skill` mentioned in the docs or in a skill body resolves |
| US English spelling | No British spellings, by exact word form |
| `## Never` blocks | Bullets inside one block agree on terminal punctuation |
+| OpenAI/Codex manifests | The second set of manifests matches the Claude ones |
| Source catalog | Structure, license vocabulary, and every skill it names resolves |
| Skill source files | Each skill's `references/sources.md` matches the catalog |
| Verticals | Every vertical emits, and the emitted repository passes its own checks |
@@ -116,6 +117,16 @@ department cannot end up missing from the README and the org chart while the che
Give it a chief before any specialists — the department's remit should exist before things are added
to it.
+## Two tools, one tree
+
+The skills install in Claude Code and in ChatGPT/Codex from the same repository. The `SKILL.md`
+format is common to both; only the manifests differ, and the second set is generated from the first
+by `scripts/build-port.py` and verified with `--check`.
+
+There is no Codex copy of the tree and there should never be one. A fork of 172 skills diverges
+within a quarter and the divergence is invisible — which is the whole argument for generating
+sixteen small manifests instead.
+
## Adding a source
`sources/*.toml` maps outside authorities — regulators, standards bodies, primary law, public
diff --git a/README.md b/README.md
index 8843e1d..43f21eb 100644
--- a/README.md
+++ b/README.md
@@ -23,19 +23,27 @@
<a href="https://cbrock84.github.io/headcount/org-chart.html"><b>Open the interactive org chart</b></a> — search every skill, open a department, jump to the source.
</p>
-An agent organization for [Claude Code](https://claude.com/claude-code), structured as a company:
-a chief executive over 16 departments, 172 skills in total.
+An agent organization structured as a company: a chief executive over
+16 departments, 172 skills in total.
Every department is an independently installable plugin, so a project loads only the functions it
needs rather than all of them at once.
## Install
+**[Claude Code](https://claude.com/claude-code)**
+
```
/plugin marketplace add cbrock84/headcount
/plugin install security@headcount
```
+**ChatGPT and Codex** — 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
+this tree, so a fix reaches both at once. See `AGENTS.md`.
+
Install as many departments as the project needs. Skills are addressed as `department:skill` —
`security:threat-modeling`, `finance:unit-economics` — so names never collide.
diff --git a/docs/AGENT-SURFACES.md b/docs/AGENT-SURFACES.md
index 2315d09..b80e1ff 100644
--- a/docs/AGENT-SURFACES.md
+++ b/docs/AGENT-SURFACES.md
@@ -155,6 +155,8 @@ sources/**
```surface:repo-meta
LICENSE
.gitignore
+AGENTS.md
+.agents/**
CONTRIBUTING.md
.gitattributes
docs/**
diff --git a/docs/DECISION-LOG.md b/docs/DECISION-LOG.md
index 84c554a..c15a281 100644
--- a/docs/DECISION-LOG.md
+++ b/docs/DECISION-LOG.md
@@ -49,6 +49,7 @@ recorded rather than being deleted.
| D35 | Whether description shape is a checkable convention | ✅ Resolved |
| D36 | Where emitted vertical repositories live | ✅ Resolved |
| D37 | Authoritative sources as a catalog of references | ✅ Resolved |
+| D38 | Running in ChatGPT as well as Claude Code | ✅ Resolved |
---
@@ -1308,3 +1309,44 @@ its body, which the waiver was tested against by pasting some in and watching it
**Coverage is `security`, `legal-risk` and `finance` first** — the three departments where citing
the authority changes the answer rather than decorating it. Extending it is adding entries, not
changing the mechanism.
+
+---
+
+## D38. Running in ChatGPT as well as Claude Code — ✅ Resolved
+
+The ask was to port the catalog so it works outside Claude Code. Checking what a port now requires
+turned out to make the decision, because the answer changed: **ChatGPT reads the same `SKILL.md`** —
+YAML frontmatter with `name` and `description`, a body, and optional `scripts/`, `references/` and
+`assets/` alongside. There is no content to port. What differs is where each tool looks for the
+manifests that describe a bundle.
+
+| | Plugin manifest | Marketplace |
+|---|---|---|
+| Claude Code | `plugins/<dept>/.claude-plugin/plugin.json` | `.claude-plugin/marketplace.json` |
+| OpenAI / Codex | `plugins/<dept>/.codex-plugin/plugin.json` | `.agents/plugins/marketplace.json` |
+
+- **(a) A second set of manifests over the same tree, generated and `--check`ed.** ← **chosen**
+- (b) Emit a standalone `headcount-openai` repository, the way `build-vertical.py` emits verticals.
+ Consistent with the machinery already built, and wrong here: a vertical emits a *different*
+ catalog, while this would emit the same 172 skills twice. Two install sources for identical
+ content, and every skill fix needing a re-emit before it reaches half the users.
+- (c) Commit a `Codex/` copy of the tree. This was offered to the repository as a contribution. It
+ is the fork D8 rejected, and the copy was a month stale before it was reviewed.
+
+**Resolution: (a).** Sixteen small manifests, a marketplace file and an `AGENTS.md`, all generated
+from the Claude manifests and the tree, all verified by `--check`. One clone installs in either
+tool, and a skill fixed once is fixed for both because there is only one copy of it.
+
+**Two things this settles beyond the mechanism.**
+
+The per-department split stops being merely tidy and becomes required. ChatGPT budgets skill
+descriptions at roughly 8,000 characters; 172 descriptions is about 50,000. Installing the whole
+catalog at once was never the intended use and on that surface it is not a possible one.
+
+`AGENTS.md` is generated rather than written, because it states counts. It is read by Codex, Cursor,
+Gemini CLI and Copilot among others, and describes the repository to an agent working *on* it —
+which is a different job from the skills, which describe how to do the work of a function.
+
+**The Codex category vocabulary is not published**, and Codex's own marketplace uses a small set of
+broad labels. Every department here is a business function, so all sixteen carry one conservative
+label rather than sixteen guesses. Revisit if a vocabulary is documented.
diff --git a/plugins/corporate-strategy/.codex-plugin/plugin.json b/plugins/corporate-strategy/.codex-plugin/plugin.json
new file mode 100644
index 0000000..0108eb9
--- /dev/null
+++ b/plugins/corporate-strategy/.codex-plugin/plugin.json
@@ -0,0 +1,25 @@
+{
+ "name": "corporate-strategy",
+ "version": "1.0.0",
+ "description": "Portfolio strategy, corporate development, strategic alliances, and scenario planning.",
+ "author": {
+ "name": "Chris Brock"
+ },
+ "repository": "https://github.com/cbrock84/headcount",
+ "license": "MIT",
+ "keywords": [
+ "strategy",
+ "corp-dev",
+ "m-and-a",
+ "partnerships",
+ "scenario-planning"
+ ],
+ "skills": "./skills/",
+ "interface": {
+ "displayName": "Corporate Strategy",
+ "category": "Productivity",
+ "shortDescription": "Corporate Strategy, as an installable department",
+ "longDescription": "Portfolio strategy, corporate development, strategic alliances, and scenario planning.",
+ "developerName": "Chris Brock"
+ }
+}
diff --git a/plugins/customer-experience/.codex-plugin/plugin.json b/plugins/customer-experience/.codex-plugin/plugin.json
new file mode 100644
index 0000000..9a77199
--- /dev/null
+++ b/plugins/customer-experience/.codex-plugin/plugin.json
@@ -0,0 +1,25 @@
+{
+ "name": "customer-experience",
+ "version": "1.0.0",
+ "description": "Support operations, escalation management, voice of customer, and self-service. Owns what the customer experiences after the sale.",
+ "author": {
+ "name": "Chris Brock"
+ },
+ "repository": "https://github.com/cbrock84/headcount",
+ "license": "MIT",
+ "keywords": [
+ "support",
+ "customer-success",
+ "escalation",
+ "voice-of-customer",
+ "service"
+ ],
+ "skills": "./skills/",
+ "interface": {
+ "displayName": "Customer Experience",
+ "category": "Productivity",
+ "shortDescription": "Customer Experience, as an installable department",
+ "longDescription": "Support operations, escalation management, voice of customer, and self-service. Owns what the customer experiences after the sale.",
+ "developerName": "Chris Brock"
+ }
+}
diff --git a/plugins/data-analytics/.codex-plugin/plugin.json b/plugins/data-analytics/.codex-plugin/plugin.json
new file mode 100644
index 0000000..491e3d9
--- /dev/null
+++ b/plugins/data-analytics/.codex-plugin/plugin.json
@@ -0,0 +1,26 @@
+{
+ "name": "data-analytics",
+ "version": "1.0.0",
+ "description": "Data governance, warehouse and semantic modeling, business intelligence, and AI/ML governance.",
+ "author": {
+ "name": "Chris Brock"
+ },
+ "repository": "https://github.com/cbrock84/headcount",
+ "license": "MIT",
+ "keywords": [
+ "data",
+ "analytics",
+ "governance",
+ "warehouse",
+ "business-intelligence",
+ "ai-governance"
+ ],
+ "skills": "./skills/",
+ "interface": {
+ "displayName": "Data & Analytics",
+ "category": "Productivity",
+ "shortDescription": "Data & Analytics, as an installable department",
+ "longDescription": "Data governance, warehouse and semantic modeling, business intelligence, and AI/ML governance.",
+ "developerName": "Chris Brock"
+ }
+}
diff --git a/plugins/demand-generation/.codex-plugin/plugin.json b/plugins/demand-generation/.codex-plugin/plugin.json
new file mode 100644
index 0000000..7d9c7b8
--- /dev/null
+++ b/plugins/demand-generation/.codex-plugin/plugin.json
@@ -0,0 +1,26 @@
+{
+ "name": "demand-generation",
+ "version": "1.0.0",
+ "description": "Organic and paid acquisition, conversion, measurement, experimentation, lifecycle messaging, and distribution.",
+ "author": {
+ "name": "Chris Brock"
+ },
+ "repository": "https://github.com/cbrock84/headcount",
+ "license": "MIT",
+ "keywords": [
+ "seo",
+ "paid-ads",
+ "cro",
+ "analytics",
+ "lifecycle",
+ "growth"
+ ],
+ "skills": "./skills/",
+ "interface": {
+ "displayName": "Demand Generation",
+ "category": "Productivity",
+ "shortDescription": "Demand Generation, as an installable department",
+ "longDescription": "Organic and paid acquisition, conversion, measurement, experimentation, lifecycle messaging, and distribution.",
+ "developerName": "Chris Brock"
+ }
+}
diff --git a/plugins/executive/.codex-plugin/plugin.json b/plugins/executive/.codex-plugin/plugin.json
new file mode 100644
index 0000000..103c03e
--- /dev/null
+++ b/plugins/executive/.codex-plugin/plugin.json
@@ -0,0 +1,23 @@
+{
+ "name": "executive",
+ "version": "1.0.0",
+ "description": "Direction, capital and attention allocation, cross-functional arbitration, strategy validation, and the method for structuring the organization itself.",
+ "author": {
+ "name": "Chris Brock"
+ },
+ "repository": "https://github.com/cbrock84/headcount",
+ "license": "MIT",
+ "keywords": [
+ "strategy",
+ "executive",
+ "decision-making"
+ ],
+ "skills": "./skills/",
+ "interface": {
+ "displayName": "Office of the CEO",
+ "category": "Productivity",
+ "shortDescription": "Office of the CEO, as an installable department",
+ "longDescription": "Direction, capital and attention allocation, cross-functional arbitration, strategy validation, and the method for structuring the organization itself.",
+ "developerName": "Chris Brock"
+ }
+}
diff --git a/plugins/finance/.codex-plugin/plugin.json b/plugins/finance/.codex-plugin/plugin.json
new file mode 100644
index 0000000..5e56c39
--- /dev/null
+++ b/plugins/finance/.codex-plugin/plugin.json
@@ -0,0 +1,24 @@
+{
+ "name": "finance",
+ "version": "1.0.0",
+ "description": "Financial modeling, budgeting and forecasting, unit economics, and financial decision support.",
+ "author": {
+ "name": "Chris Brock"
+ },
+ "repository": "https://github.com/cbrock84/headcount",
+ "license": "MIT",
+ "keywords": [
+ "finance",
+ "budgeting",
+ "forecasting",
+ "unit-economics"
+ ],
+ "skills": "./skills/",
+ "interface": {
+ "displayName": "Finance",
+ "category": "Productivity",
+ "shortDescription": "Finance, as an installable department",
+ "longDescription": "Financial modeling, budgeting and forecasting, unit economics, and financial decision support.",
+ "developerName": "Chris Brock"
+ }
+}
diff --git a/plugins/it-operations/.codex-plugin/plugin.json b/plugins/it-operations/.codex-plugin/plugin.json
new file mode 100644
index 0000000..db3ccac
--- /dev/null
+++ b/plugins/it-operations/.codex-plugin/plugin.json
@@ -0,0 +1,28 @@
+{
+ "name": "it-operations",
+ "version": "1.0.0",
+ "description": "Corporate IT: service desk, systems and network administration, virtualization and cloud, telephony and conferencing, endpoints, assets, identity lifecycle, and backup and recovery.",
+ "author": {
+ "name": "Chris Brock"
+ },
+ "repository": "https://github.com/cbrock84/headcount",
+ "license": "MIT",
+ "keywords": [
+ "it-operations",
+ "service-desk",
+ "sysadmin",
+ "network",
+ "endpoint-management",
+ "virtualization",
+ "cloud",
+ "telephony"
+ ],
+ "skills": "./skills/",
+ "interface": {
+ "displayName": "IT Operations",
+ "category": "Productivity",
+ "shortDescription": "IT Operations, as an installable department",
+ "longDescription": "Corporate IT: service desk, systems and network administration, virtualization and cloud, telephony and conferencing, endpoints, assets, identity lifecycle, and backup and recovery.",
+ "developerName": "Chris Brock"
+ }
+}
diff --git a/plugins/legal-risk/.codex-plugin/plugin.json b/plugins/legal-risk/.codex-plugin/plugin.json
new file mode 100644
index 0000000..0bc5547
--- /dev/null
+++ b/plugins/legal-risk/.codex-plugin/plugin.json
@@ -0,0 +1,25 @@
+{
+ "name": "legal-risk",
+ "version": "1.0.0",
+ "description": "Contract review, privacy and data protection, enterprise risk, corporate governance, compliance, and audit readiness. Reviewer-class: its findings are not overrulable by the department under review.",
+ "author": {
+ "name": "Chris Brock"
+ },
+ "repository": "https://github.com/cbrock84/headcount",
+ "license": "MIT",
+ "keywords": [
+ "legal",
+ "privacy",
+ "risk",
+ "compliance",
+ "contracts"
+ ],
+ "skills": "./skills/",
+ "interface": {
+ "displayName": "Legal & Risk",
+ "category": "Productivity",
+ "shortDescription": "Legal & Risk, as an installable department",
+ "longDescription": "Contract review, privacy and data protection, enterprise risk, corporate governance, compliance, and audit readiness. Reviewer-class: its findings are not overrulable by the department under review.",
+ "developerName": "Chris Brock"
+ }
+}
diff --git a/plugins/marketing/.codex-plugin/plugin.json b/plugins/marketing/.codex-plugin/plugin.json
new file mode 100644
index 0000000..75c1616
--- /dev/null
+++ b/plugins/marketing/.codex-plugin/plugin.json
@@ -0,0 +1,25 @@
+{
+ "name": "marketing",
+ "version": "1.0.0",
+ "description": "Positioning, content strategy and production, social and video, brand voice, copywriting, PR, and partnerships.",
+ "author": {
+ "name": "Chris Brock"
+ },
+ "repository": "https://github.com/cbrock84/headcount",
+ "license": "MIT",
+ "keywords": [
+ "marketing",
+ "content",
+ "brand",
+ "social",
+ "communications"
+ ],
+ "skills": "./skills/",
+ "interface": {
+ "displayName": "Marketing",
+ "category": "Productivity",
+ "shortDescription": "Marketing, as an installable department",
+ "longDescription": "Positioning, content strategy and production, social and video, brand voice, copywriting, PR, and partnerships.",
+ "developerName": "Chris Brock"
+ }
+}
diff --git a/plugins/operations/.codex-plugin/plugin.json b/plugins/operations/.codex-plugin/plugin.json
new file mode 100644
index 0000000..38633bd
--- /dev/null
+++ b/plugins/operations/.codex-plugin/plugin.json
@@ -0,0 +1,24 @@
+{
+ "name": "operations",
+ "version": "1.0.0",
+ "description": "Program management, process design, vendor and supplier management, and operational delivery.",
+ "author": {
+ "name": "Chris Brock"
+ },
+ "repository": "https://github.com/cbrock84/headcount",
+ "license": "MIT",
+ "keywords": [
+ "operations",
+ "process",
+ "program-management",
+ "vendors"
+ ],
+ "skills": "./skills/",
+ "interface": {
+ "displayName": "Operations",
+ "category": "Productivity",
+ "shortDescription": "Operations, as an installable department",
+ "longDescription": "Program management, process design, vendor and supplier management, and operational delivery.",
+ "developerName": "Chris Brock"
+ }
+}
diff --git a/plugins/people/.codex-plugin/plugin.json b/plugins/people/.codex-plugin/plugin.json
new file mode 100644
index 0000000..c421aab
--- /dev/null
+++ b/plugins/people/.codex-plugin/plugin.json
@@ -0,0 +1,25 @@
+{
+ "name": "people",
+ "version": "1.0.0",
+ "description": "Org design, hiring and interviewing, compensation and leveling, and workforce planning.",
+ "author": {
+ "name": "Chris Brock"
+ },
+ "repository": "https://github.com/cbrock84/headcount",
+ "license": "MIT",
+ "keywords": [
+ "people",
+ "hr",
+ "org-design",
+ "hiring",
+ "compensation"
+ ],
+ "skills": "./skills/",
+ "interface": {
+ "displayName": "People",
+ "category": "Productivity",
+ "shortDescription": "People, as an installable department",
+ "longDescription": "Org design, hiring and interviewing, compensation and leveling, and workforce planning.",
+ "developerName": "Chris Brock"
+ }
+}
diff --git a/plugins/pmo/.codex-plugin/plugin.json b/plugins/pmo/.codex-plugin/plugin.json
new file mode 100644
index 0000000..d744413
--- /dev/null
+++ b/plugins/pmo/.codex-plugin/plugin.json
@@ -0,0 +1,25 @@
+{
+ "name": "pmo",
+ "version": "1.0.0",
+ "description": "Enterprise program management office: portfolio governance, program and project delivery, dependencies and delivery risk, benefits realization, and adoption.",
+ "author": {
+ "name": "Chris Brock"
+ },
+ "repository": "https://github.com/cbrock84/headcount",
+ "license": "MIT",
+ "keywords": [
+ "pmo",
+ "portfolio",
+ "project-management",
+ "program-management",
+ "delivery"
+ ],
+ "skills": "./skills/",
+ "interface": {
+ "displayName": "Program Management Office",
+ "category": "Productivity",
+ "shortDescription": "Program Management Office, as an installable department",
+ "longDescription": "Enterprise program management office: portfolio governance, program and project delivery, dependencies and delivery risk, benefits realization, and adoption.",
+ "developerName": "Chris Brock"
+ }
+}
diff --git a/plugins/product/.codex-plugin/plugin.json b/plugins/product/.codex-plugin/plugin.json
new file mode 100644
index 0000000..2191db9
--- /dev/null
+++ b/plugins/product/.codex-plugin/plugin.json
@@ -0,0 +1,25 @@
+{
+ "name": "product",
+ "version": "1.0.0",
+ "description": "Product strategy, UX and usability audit, design systems, interface craft, brand expression, and visual direction.",
+ "author": {
+ "name": "Chris Brock"
+ },
+ "repository": "https://github.com/cbrock84/headcount",
+ "license": "MIT",
+ "keywords": [
+ "product",
+ "ux",
+ "design",
+ "design-system",
+ "brand"
+ ],
+ "skills": "./skills/",
+ "interface": {
+ "displayName": "Product",
+ "category": "Productivity",
+ "shortDescription": "Product, as an installable department",
+ "longDescription": "Product strategy, UX and usability audit, design systems, interface craft, brand expression, and visual direction.",
+ "developerName": "Chris Brock"
+ }
+}
diff --git a/plugins/revenue/.codex-plugin/plugin.json b/plugins/revenue/.codex-plugin/plugin.json
new file mode 100644
index 0000000..a6a2b26
--- /dev/null
+++ b/plugins/revenue/.codex-plugin/plugin.json
@@ -0,0 +1,25 @@
+{
+ "name": "revenue",
+ "version": "1.0.0",
+ "description": "Outbound, sales enablement, revenue operations, pricing and packaging, activation, retention, and referrals.",
+ "author": {
+ "name": "Chris Brock"
+ },
+ "repository": "https://github.com/cbrock84/headcount",
+ "license": "MIT",
+ "keywords": [
+ "sales",
+ "pricing",
+ "revenue",
+ "retention",
+ "revops"
+ ],
+ "skills": "./skills/",
+ "interface": {
+ "displayName": "Revenue",
+ "category": "Productivity",
+ "shortDescription": "Revenue, as an installable department",
+ "longDescription": "Outbound, sales enablement, revenue operations, pricing and packaging, activation, retention, and referrals.",
+ "developerName": "Chris Brock"
+ }
+}
diff --git a/plugins/security/.codex-plugin/plugin.json b/plugins/security/.codex-plugin/plugin.json
new file mode 100644
index 0000000..07207cc
--- /dev/null
+++ b/plugins/security/.codex-plugin/plugin.json
@@ -0,0 +1,26 @@
+{
+ "name": "security",
+ "version": "1.0.0",
+ "description": "Threat modeling, security architecture review, incident response, vulnerability management, and access and identity. Reviewer-class: blocking findings are not overrulable by the department under review.",
+ "author": {
+ "name": "Chris Brock"
+ },
+ "repository": "https://github.com/cbrock84/headcount",
+ "license": "MIT",
+ "keywords": [
+ "security",
+ "appsec",
+ "threat-modeling",
+ "incident-response",
+ "vulnerability-management",
+ "identity"
+ ],
+ "skills": "./skills/",
+ "interface": {
+ "displayName": "Security",
+ "category": "Productivity",
+ "shortDescription": "Security, as an installable department",
+ "longDescription": "Threat modeling, security architecture review, incident response, vulnerability management, and access and identity. Reviewer-class: blocking findings are not overrulable by the department under review.",
+ "developerName": "Chris Brock"
+ }
+}
diff --git a/plugins/technology/.codex-plugin/plugin.json b/plugins/technology/.codex-plugin/plugin.json
new file mode 100644
index 0000000..2587188
--- /dev/null
+++ b/plugins/technology/.codex-plugin/plugin.json
@@ -0,0 +1,25 @@
+{
+ "name": "technology",
+ "version": "1.0.0",
+ "description": "Architecture, engineering delivery, development discipline, debugging, review, and AI and automation systems.",
+ "author": {
+ "name": "Chris Brock"
+ },
+ "repository": "https://github.com/cbrock84/headcount",
+ "license": "MIT",
+ "keywords": [
+ "engineering",
+ "tdd",
+ "debugging",
+ "code-review",
+ "automation"
+ ],
+ "skills": "./skills/",
+ "interface": {
+ "displayName": "Technology",
+ "category": "Productivity",
+ "shortDescription": "Technology, as an installable department",
+ "longDescription": "Architecture, engineering delivery, development discipline, debugging, review, and AI and automation systems.",
+ "developerName": "Chris Brock"
+ }
+}
diff --git a/scripts/build-port.py b/scripts/build-port.py
new file mode 100755
index 0000000..d9a564b
--- /dev/null
+++ b/scripts/build-port.py
@@ -0,0 +1,203 @@
+#!/usr/bin/env python3
+"""Generate the OpenAI/Codex manifests so this repository installs in ChatGPT as well as Claude.
+
+The skills need no porting. ChatGPT and Claude Code read the same `SKILL.md` — YAML frontmatter
+with `name` and `description`, a body, and optional `scripts/`, `references/` and `assets/`
+alongside. What differs is where each tool looks for the manifests that describe a bundle:
+
+ Claude Code plugins/<dept>/.claude-plugin/plugin.json .claude-plugin/marketplace.json
+ OpenAI/Codex plugins/<dept>/.codex-plugin/plugin.json .agents/plugins/marketplace.json
+
+So the port is a second set of manifests over the same tree, not a second tree. That is why this
+emits *into* the repository rather than into `dist/` the way `build-vertical.py` does: there is no
+fork to keep in step, one clone installs in either tool, and a skill fixed once is fixed for both.
+
+The alternative, which this repository already had proposed to it, is to copy the whole tree into a
+`Codex/` directory. That is the fork D8 rejected, and the copy offered was a month stale before it
+was reviewed. Sixteen small generated manifests do the same job and cannot drift.
+
+Generated, never hand-edited: `--check` fails the build when a manifest drifts from its Claude
+counterpart.
+"""
+import argparse
+import glob
+import json
+import os
+import re
+import sys
+
+ROOT = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
+
+# Display titles live in build-readme.py's META and nowhere else. Parsed rather than imported,
+# because importing that module runs its tree walk as a side effect; parsed rather than duplicated,
+# because a second copy of sixteen titles is a second thing to forget.
+META_ENTRY = re.compile(r'^\s*"([a-z-]+)":\s*\(\s*\d+,\s*"([^"]+)",\s*"([^"]+)"\s*\),', re.M)
+
+# Codex publishes no category vocabulary, and its own marketplace entries use a small set of broad
+# labels. Everything here is a business function, so one conservative label beats sixteen guesses.
+CATEGORY = "Productivity"
+
+
+def titles():
+ source = open(os.path.join(ROOT, "scripts", "build-readme.py"), encoding="utf-8").read()
+ found = {slug: title for slug, title, _ in META_ENTRY.findall(source)}
+ if not found:
+ sys.exit("build-port: could not read META from scripts/build-readme.py")
+ return found
+
+
+def departments(marketplace):
+ return [p["name"] for p in marketplace["plugins"]]
+
+
+def codex_plugin(claude, title):
+ """The Codex manifest for one department, from the Claude manifest of the same department."""
+ return {
+ "name": claude["name"],
+ "version": claude["version"],
+ "description": claude["description"],
+ "author": claude["author"],
+ "repository": claude.get("repository", ""),
+ "license": "MIT",
+ "keywords": claude.get("keywords", []),
+ # A path, not a list — Codex discovers the directory. One less thing to regenerate when a
+ # skill is added.
+ "skills": "./skills/",
+ "interface": {
+ "displayName": title,
+ "category": CATEGORY,
+ "shortDescription": f"{title}, as an installable department",
+ "longDescription": claude["description"],
+ "developerName": claude["author"]["name"],
+ },
+ }
+
+
+def agents_md(department_count, skill_count, titles_by_slug, order):
+ """The cross-tool repository context file.
+
+ AGENTS.md is read by Codex, Cursor, Gemini CLI, Copilot and others, and describes the repository
+ to an agent working *in* it — which is a different job from the skills, which describe how to do
+ business work. Generated so the counts cannot go stale.
+ """
+ lines = [
+ "# AGENTS.md",
+ "",
+ "<!-- Generated by scripts/build-port.py. Do not edit. -->",
+ "",
+ f"An agent organization structured as a company: {department_count} departments, "
+ f"{skill_count} skills, each department installable on its own.",
+ "",
+ "This file describes the repository to an agent working *on* it. The skills under "
+ "`plugins/` are a different thing — they describe how to do the work of a function, and "
+ "they load when a request matches their description.",
+ "",
+ "## Installing, in either tool",
+ "",
+ "The skills are identical. Only the manifests differ, and both are generated from the same "
+ "tree, so a skill fixed once is fixed for both.",
+ "",
+ "| | Manifests |",
+ "|---|---|",
+ "| Claude Code | `plugins/<department>/.claude-plugin/plugin.json`, `.claude-plugin/marketplace.json` |",
+ "| ChatGPT / Codex | `plugins/<department>/.codex-plugin/plugin.json`, `.agents/plugins/marketplace.json` |",
+ "",
+ "## Before you open a pull request",
+ "",
+ "```",
+ "./scripts/check-all.sh",
+ "```",
+ "",
+ "Every check CI runs, in one script — the workflow calls this same file, so the two cannot "
+ "drift. Stage new files first: the surface guard reads `git ls-files`, so an unstaged file "
+ "is invisible to it and the check passes locally then fails in CI.",
+ "",
+ "## Rules that are enforced rather than suggested",
+ "",
+ "- **One owner per path.** `docs/AGENT-SURFACES.md` maps every tracked path to exactly one "
+ "agent, and no two owners may claim the same path. A new department adds its roster row, "
+ "its surface block and its charter in the same change.",
+ "- **Generated files are never hand-edited.** The README, the org chart, the social card, "
+ "each skill's `references/sources.md`, and these manifests are all emitted from the tree "
+ "and verified with `--check`. An edit to one is lost on the next regenerate.",
+ "- **US English**, by exact word form. `scripts/check-us-english.py --fix` rewrites.",
+ "- **No third-party licensed content.** All prose here is original; "
+ "`scripts/check-provenance.py` is the backstop.",
+ "",
+ "## Departments",
+ "",
+ ]
+ lines += [f"- `{slug}` — {titles_by_slug[slug]}" for slug in order]
+ lines += [""]
+ return "\n".join(lines)
+
+
+def build():
+ marketplace = json.load(open(os.path.join(ROOT, ".claude-plugin", "marketplace.json"),
+ encoding="utf-8"))
+ titles_by_slug = titles()
+ order = departments(marketplace)
+
+ wanted = {}
+ total = 0
+ for department in order:
+ claude_path = f"plugins/{department}/.claude-plugin/plugin.json"
+ claude = json.load(open(os.path.join(ROOT, claude_path), encoding="utf-8"))
+ total += len(glob.glob(os.path.join(ROOT, "plugins", department, "skills", "*", "SKILL.md")))
+ wanted[f"plugins/{department}/.codex-plugin/plugin.json"] = json.dumps(
+ codex_plugin(claude, titles_by_slug[department]), indent=2) + "\n"
+
+ wanted[".agents/plugins/marketplace.json"] = json.dumps({
+ "name": marketplace["name"],
+ "interface": {"displayName": marketplace["name"]},
+ "plugins": [
+ {
+ "name": p["name"],
+ "source": {"source": "local", "path": f"./plugins/{p['name']}"},
+ "policy": {"installation": "AVAILABLE"},
+ "category": CATEGORY,
+ }
+ for p in marketplace["plugins"]
+ ],
+ }, indent=2) + "\n"
+
+ wanted["AGENTS.md"] = agents_md(len(order), total, titles_by_slug, order)
+ return wanted
+
+
+def main():
+ parser = argparse.ArgumentParser(description=__doc__)
+ parser.add_argument("--check", action="store_true", help="report drift instead of writing")
+ args = parser.parse_args()
+
+ wanted = build()
+ # A department removed from the tree must lose its Codex manifest too, or the marketplace
+ # advertises a plugin that is not there.
+ stale = sorted(set(glob.glob("plugins/*/.codex-plugin/plugin.json")) - set(wanted))
+
+ if args.check:
+ drifted = [p for p, text in sorted(wanted.items())
+ if not os.path.exists(p) or open(p, encoding="utf-8").read() != text]
+ for path in drifted:
+ print(f" stale or missing: {path}")
+ for path in stale:
+ print(f" department no longer exists: {path}")
+ problems = len(drifted) + len(stale)
+ if problems:
+ print(" run: python3 scripts/build-port.py")
+ print(f"openai/codex manifests: {len(wanted)} expected, {problems} out of date")
+ return 1 if problems else 0
+
+ for path, text in sorted(wanted.items()):
+ os.makedirs(os.path.dirname(path) or ".", exist_ok=True)
+ with open(path, "w", encoding="utf-8") as handle:
+ handle.write(text)
+ for path in stale:
+ os.remove(path)
+ print(f"openai/codex manifests: {len(wanted)} written"
+ + (f", {len(stale)} removed" if stale else ""))
+ return 0
+
+
+if __name__ == "__main__":
+ sys.exit(main())
diff --git a/scripts/build-readme.py b/scripts/build-readme.py
index 55e2fc5..eb38162 100755
--- a/scripts/build-readme.py
+++ b/scripts/build-readme.py
@@ -99,19 +99,27 @@ out = [
" chart</b></a> — search every skill, open a department, jump to the source.",
"</p>",
"",
- "An agent organization for [Claude Code](https://claude.com/claude-code), structured as a company:",
- f"a chief executive over {len(ORDER)} departments, {total} skills in total.",
+ "An agent organization structured as a company: a chief executive over",
+ f"{len(ORDER)} departments, {total} skills in total.",
"",
"Every department is an independently installable plugin, so a project loads only the functions it",
"needs rather than all of them at once.",
"",
"## Install",
"",
+ "**[Claude Code](https://claude.com/claude-code)**",
+ "",
"```",
"/plugin marketplace add cbrock84/headcount",
"/plugin install security@headcount",
"```",
"",
+ "**ChatGPT and Codex** — 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",
+ "this tree, so a fix reaches both at once. See `AGENTS.md`.",
+ "",
"Install as many departments as the project needs. Skills are addressed as `department:skill` —",
"`security:threat-modeling`, `finance:unit-economics` — so names never collide.",
"",
diff --git a/scripts/check-all.sh b/scripts/check-all.sh
index 329526c..ffa454a 100755
--- a/scripts/check-all.sh
+++ b/scripts/check-all.sh
@@ -35,6 +35,8 @@ run "US English spelling" \
python3 scripts/check-us-english.py
run "Never blocks are internally consistent" \
python3 scripts/check-never-blocks.py
+run "OpenAI/Codex manifests are current" \
+ python3 scripts/build-port.py --check
run "Source catalog is valid" \
python3 scripts/check-sources.py
run "Skill source files are current" \
@@ -45,7 +47,7 @@ run "Manifests parse" \
python3 -c "
import json,glob,sys
bad=[]
-for f in ['.claude-plugin/marketplace.json']+glob.glob('plugins/*/.claude-plugin/plugin.json'):
+for f in ['.claude-plugin/marketplace.json','.agents/plugins/marketplace.json']+glob.glob('plugins/*/.claude-plugin/plugin.json')+glob.glob('plugins/*/.codex-plugin/plugin.json'):
try: json.load(open(f))
except Exception as e: bad.append(f'{f}: {e}')
for b in bad: print(' ',b)