ai-company

An LLM Agent governance skill — micro tier: 2 departments, 18 function blocks, one installable skill package.

English | 简体中文

README.en.md is an identical English copy of this file, kept so the .en / .zh language suffixes stay symmetric for i18n tooling.

Quick Start

1. Install — copy the ai-company/ folder into your agent's skills workspace.
2. Activate — the skill registers through its single runtime entry point, SKILL.md; no other file participates in routing or activation.
3. Use — ask your agent a governance or delivery question; it routes via SKILL.md to the relevant department specification under references/departments/.
4. Human mode — copy a ready-made prompt from prompts/01-implement-method.md or prompts/02-robustness-checks.md into any AI chat window. Both files are human-paste mode and reference no internal paths.
5. Verify — run python tests/test-method-patterns.py, and check tier readiness with powershell -File scripts/self-scale.ps1 -Action evaluate.

Package at a glance

Item	Value
Skill name	ai-company
Version	1.0.0
Scale tier	micro (XS)
Departments	2 (governance-and-delivery, engineering-and-safety)
Function blocks	18 (8 + 10)
Total files	26
License	GPL-3.0

Project Structure

ai-company/
├── .editorconfig
├── .gitattributes
├── .gitignore
├── .scaling-state.json
├── AGENTS.md
├── CHANGELOG.md
├── CODE_OF_CONDUCT.md
├── CONTRIBUTING.md
├── LICENSE
├── README-FOR-AI.md
├── README.en.md
├── README.md
├── README.zh.md
├── SECURITY.md
├── SKILL.md
├── _meta.json
├── prompts/
│   ├── 01-implement-method.md
│   └── 02-robustness-checks.md
├── references/
│   ├── method-patterns.md
│   ├── error-codes.md
│   ├── scaling.md
│   └── departments/
│       ├── governance-and-delivery.md
│       └── engineering-and-safety.md
├── scripts/
│   ├── self-scale.ps1
│   └── scaling-config.json
└── tests/
    └── test-method-patterns.py

Documentation

• CONTRIBUTING.md — branch, commit, test, and PR rules
• SECURITY.md — supported versions and vulnerability reporting
• CHANGELOG.md — release history
• CODE_OF_CONDUCT.md — community standards
• AGENTS.md — entry point for coding agents maintaining this repository (does not participate in skill routing)
• README-FOR-AI.md — generation & maintenance spec for LLM Agents (self-contained and authoritative for this standalone project; no external spec is required)

Upgrading

This package is designed to grow. scripts/self-scale.ps1 -Action evaluate checks tier readiness against scripts/scaling-config.json and writes back only its own bookkeeping fields to .scaling-state.json (next_evaluation, last_metrics, routing_miss_streak); it never modifies skill content. The actual upgrade always requires human approval and is recorded in .scaling-state.json. Thresholds, paths, and aliases are specified in references/scaling.md; the scale-tier ladder itself is defined locally in README-FOR-AI.md §1.

License

Copyright (c) 2026. Licensed under the GNU General Public License v3.0.