company-init
Skill8 starsChanged 3 months agoMITView on GitHub
Found or adopt the company - greenfield spec and architecture, or an existing-codebase audit that wires real gates and frozen defaults. The owner answers at most one question. Use when the user says /company-init or /onboard, or asks to set up claude-company in a fresh or existing repo. /company self-initializes too; this skill is the explicit founding or adoption pass.
In this repository4 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- brainstormSkill
- companySkill
- lean-companySkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 18
If the request already says what we are building, ask nothing.
- Very specific: line 39
Ask nothing. Audit in parallel (architecture, tribal conventions, real test/lint/build commands, load-bearing surfaces).
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
/company-init - found or adopt
The owner states what they want. The company generates the rest. Output is
a project where /company runs real work.
0. Verify the drop-in
company/METHOD.md, COMPANY.md, .claude/agents/, .claude/hooks/
must exist (installer). If not: stop, point at install.sh.
Empty repo - found it
If the request already says what we are building, ask nothing. Otherwise one question: "What are we building, and for whom?"
Then, in order:
- You write the founding spec in
company/specs/with FR/BR IDs and fallbacks, with the owner. If the client wants options first,/brainstormalready ran. - architect: the landed waist, ownership map, wave plan, frozen-surface entries, gate proposal.
- You wire:
company/frozen-surfaces.json,company/gates.config(python3 .claude/hooks/gates_detect.py --writeonce code exists; until then wirepython3 .claude/hooks/guard_models.py --checkso founding commits are honest), spawn facts in RESUME, conventions inCLAUDE.md,git initif needed.
Report one screen: what you decided, what they can veto. Then roll into
/company if they already asked to build.
Existing codebase - adopt it
Ask nothing. Audit in parallel (architecture, tribal conventions, real test/lint/build commands, load-bearing surfaces). Then:
python3 .claude/hooks/gates_detect.py --write, reconcile with CI, runbash company/run-gates.sh. Red-today is debt in RESUME, not a silent pass.- Freeze shipped migrations, schema, single-writer files. Owner can veto.
- Extend
CLAUDE.md(never clobber) from evidence. - Seed RESUME and WORRIES.
Findings memo, one screen: what you wired, current gate colors, worries.
Then /company is open.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

