Contract Agents: 125 AI Specialists That Actually Follow the Rules
You ask an AI agent to refactor a module. It rewrites half the codebase. You ask it to fix a bug. It adds three features you didn't request. You ask it to deploy. It force-pushes to main.
AI agents have a discipline problem. Contract Agents is the fix.
The Problem
AI coding agents are getting powerful fast. But power without discipline is a liability. Here's what actually happens in practice:
- Scope creep — Ask for a one-line fix, get a 500-line refactor
- Inconsistency — Same task, different agent, wildly different approach
- Silent risk — Destructive operations without asking, secrets committed, hooks bypassed
- No verification — "Done!" (nothing was tested)
- Amnesia — Corrections from yesterday are forgotten today
The root cause? Every agent operates in a vacuum. No shared rules, no accountability, no structure. It's like hiring 125 contractors and giving none of them an employee handbook.
The Solution: One Contract to Rule Them All
Contract Agents is a system of 125 specialized AI agents across 10 divisions — all governed by a single shared contract: AGENTS_CONTRACT.md.
┌─────────────────────────────────────────────┐
│ CONTRACT-AGENTS │
│ │
│ 125 AI Agents • 10 Divisions • 1 Contract │
│ │
│ eng-* test-* design-* mkt-* prod-* │
│ pm-* game-* spatial-* spec-* support- │
└─────────────────────────────────────────────┘
The contract isn't a suggestion — it's the employee handbook. Every agent loads it before doing anything. It enforces:
| Rule | What It Prevents |
|---|---|
| Scope Discipline | Touch only what's required — no drive-by refactors |
| Confusion Management | Stop on ambiguity, ask clarifying questions |
| Evidence Rules | Claims require proof — file:line for code, URLs for research |
| Approval Gates | Hard stops before destructive operations |
| Verification Chain | Automated → tests → full → manual — in that order |
| Commit Discipline | Safe, bisectable git history |
| Self-Improvement | Agents learn from your corrections |
Think of It Like a Company
You're the executive. You set direction. The agents do the work.
| Component | Role |
|---|---|
| You | Executive — sets goals and direction |
AGENTS_CONTRACT.md | Employee handbook — how everyone works |
contract-orchestrator | Project manager — coordinates multi-agent workflows |
pm-senior | Strategic advisor — helps you plan |
Division agents (eng-*, test-*, etc.) | Specialists you delegate to |
The orchestrator doesn't just start working — it presents a proposed pipeline (which agents, in what order, and why) and waits for your approval before delegating anything.
10 Divisions, 125 Specialists
Each agent is minimal by design. Domain-specific rules only — everything shared lives in the contract. No bloat, no duplication.
| Division | Prefix | Count | Examples |
|---|---|---|---|
| Engineering | eng-* | 23 | frontend, backend, devops, security, mobile, AI/ML |
| Testing & QA | test-* | 11 | penetration testing, API testing, accessibility, Docker security |
| Design | design-* | 8 | UI, UX research, brand, visual storytelling |
| Marketing | mkt-* | 17 | SEO, growth hacking, social media, content |
| Product | prod-* | 5 | competitive intel, feedback synthesis, sprint planning |
| Project Management | pm-* | 7 | Jira workflows, studio ops, indie business strategy |
| Game Dev | game-* | 19 | Unity, Unreal, Godot, Roblox, narrative, audio |
| Spatial Computing | spatial-* | 6 | visionOS, WebXR, Metal, spatial interfaces |
| Specialized | spec-* | 22 | forensics, blockchain audit, compliance, architecture |
| Support | support-* | 6 | legal, finance, infrastructure, customer support |
What Makes This Different
Most "agent" projects are single mega-prompts. They cram everything into one system prompt and hope for the best. Contract Agents takes the opposite approach:
- Separation of concerns — Each agent has one job. The contract handles shared behavior.
- Agent files are tiny — Shared rules live in the contract, not duplicated across 125 files.
- Cross-tool compatibility — Same agents work in Claude Code, Kilo CLI, Goose, and OpenCode.
- Human-guided — Nothing happens without your direction. Agents propose, you approve.
Quick Start
# Clone and install globally
git clone https://github.com/gl0bal01/contract-agents.git
cd contract-agents
make install
# Or install to a specific tool
make install-claude
make install-kilo
make install-goose
make install-opencode
# Or install per-project
make install SCOPE=local LOCAL_PATH=~/my-project
# Or install just one division
make install-claude DIVISION=eng
Manual install is just as simple — copy the agent markdown files to your tool's agents directory:
cp agents/*.md ~/.claude/agents/ # Claude Code
cp agents/*.md ~/.kilo/agents/ # Kilo CLI
cp agents/*.md ~/.goose/agents/ # Goose
Usage
Don't know where to start? Three options:
# Let the orchestrator figure it out
"Use contract-orchestrator to coordinate building a new API service"
# Get strategic direction first
"Use pm-senior to plan this feature and identify which agents to use"
# Direct delegation when you know what you need
"Use eng-backend-architect to design the database schema"
"Use test-penetration-tester to audit the auth system"
"Use mkt-seo-strategist to optimize the landing page"
The Bottom Line
AI agents without rules are liabilities. AI agents with a shared contract are a workforce.
125 specialists. 10 divisions. 1 contract. You set the direction — they do the work. That's the deal.
Open source (MIT). Works with Claude Code, Kilo CLI, Goose, and OpenCode. github.com/gl0bal01/contract-agents