Corp
Agentic skills; for hako code and other agents.
npx -y skills add mithraeums/skills --skill corpAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
A company in a folder — a referential multi-agent org for any AI coding assistant. Loads CEO, DEV, DESIGN, QA, and ARCH agents from markdown on demand. Use when a task spans multiple domains, before building anything non-trivial, or to decide which agent handles a given task. Host-agnostic (Claude / Codex / Cursor / Gemini / hako / any). Trigger: /corp, "which agent", "load corp", "who handles X".
SKILL.md
2.7 KB, 625 tokens by cl100k_base, as published. Nobody here has run it
corp — company in a folder
What this is
Most skills teach your assistant how to do a task. corp teaches it who to ask. Five markdown agents — a CEO who says no, a dev lead who reads before writing, a designer who matches existing patterns, a QA who blocks bad ships, an architect who owns the system shape. No runtime, no framework, no dependencies. Just files any LLM can read.
Host adaptation
Works in any agent. The dispatcher ships as both CLAUDE.md and AGENTS.md —
your host auto-loads whichever it knows. Use your own native file/shell tools;
ignore any tools: frontmatter your host doesn't support. Output tags are plain text.
Agent map
| Agent | Domain | Call when |
|---|---|---|
agents/CEO.md | Scope, priorities, what not to build | Before building anything non-trivial |
agents/DEV.md | Code, bugs, backend, database | Writing or fixing anything |
agents/DESIGN.md | UI, components, styles, interactions | Touching the frontend |
agents/QA.md | Quality gate, ship/no-ship | Before declaring done — always |
agents/ARCH.md | Schema, API contracts, system shape | Before touching the data layer |
Quick dispatch
- "Should I build this?" → CEO
- "How do I implement X?" → DEV
- "Is this UI consistent?" → DESIGN
- "Is this ready to ship?" → QA
- "Will this break the schema?" → ARCH
Protocol
Task arrives
→ Read the dispatcher (CLAUDE.md / AGENTS.md)
→ Identify domain → load that agent
→ Agent may hand off → follow it
→ Implement
→ QA checklist
→ Ship
Setup check
If the dispatcher's [PROJECT] block is empty, fill it before using corp agents
(Name, Stack, Structure required). Run the corp-init skill to do this interactively.
Token rules
- Load agents on demand, never all at once
- Tag output:
[CEO][DEV][DESIGN][QA][ARCH] - Prefer targeted edits over rewrites
Install
Any agent (universal): copy the corp folder into your repo, then add one line to
your assistant's entry file (CLAUDE.md / AGENTS.md / system prompt):
For multi-domain tasks, read
corp/CLAUDE.md(orAGENTS.md) first.
Claude Code (plugin, adds /corp commands):
claude plugin install github:mithraeums/skills
hako / hako-code: drop the corp folder into your skills dir; the agent loads it on relevance.