agenteazy 0.2.0

Turn any GitHub repo into an AI agent in one command

AgentEazy

Turn any GitHub repo into an interoperable AI agent in one command.

Open source · MIT Licensed · agenteazy.com

---

Quick Start

pip install agenteazy
agenteazy signup <your-github-username>
agenteazy deploy github.com/you/your-repo

Your repo is now a live agent. Call it:

curl -X POST https://simondusable--agenteazy-gateway-serve.modal.run/agent/your-repo/ \
  -H "Content-Type: application/json" \
  -d '{"verb":"DO","payload":{"data":{"input":"hello"}}}'

---

Python SDK

pip install agenteazy

from agenteazy import AgentEazy

client = AgentEazy()

# Search for agents
agents = client.find("password strength")

# Call an agent
result = client.do("zxcvbn-python", {"password": "test123"})
print(result)

---

LangChain Integration

pip install agenteazy[langchain]

from agenteazy.integrations.langchain import AgentEazyTool, AgentEazyToolkit

# Single tool from one agent
tool = AgentEazyTool.from_agent("zxcvbn-python")
result = tool.run({"password": "test123"})

# Toolkit — multiple tools from a registry search
toolkit = AgentEazyToolkit()
tools = toolkit.get_tools(query="text processing", limit=5)

---

CrewAI Integration

pip install agenteazy[crewai]

from agenteazy.integrations.crewai import AgentEazyCrewTool

tool = AgentEazyCrewTool.from_agent("zxcvbn-python")
result = tool.run({"password": "test123"})

---

CLI Commands

Core

Command	Description
agenteazy analyze <repo>	Detect language, deps, entry point
agenteazy wrap <repo>	Generate agent.json + FastAPI wrapper
agenteazy wrap <repo> --entry func	Specify entry function
agenteazy wrap <repo> --env KEY=VAL	Set environment variables
agenteazy deploy <repo>	Deploy to gateway + register
agenteazy deploy <repo> --local	Test locally on port 8000
agenteazy deploy <repo> --price 10	Deploy as paid agent (10 credits/call)

Batch

Command	Description
agenteazy batch-analyze <dir>	Analyze all repos in a directory
agenteazy batch-deploy <dir>	Deploy multiple repos at once
agenteazy batch-deploy <dir> --wrap-only	Wrap without deploying
agenteazy batch-deploy <dir> --dry-run	Preview without changes
agenteazy batch-deploy <dir> --entry-file main.py	Override entry file
agenteazy batch-deploy <dir> --skip-existing	Skip already-deployed agents
agenteazy batch-deploy <dir> --max-failures 3	Stop after N failures

Account

Command	Description
agenteazy signup <username>	Create account, get API key
agenteazy balance	Check credit balance
agenteazy transactions	View transaction history

Registry

Command	Description
agenteazy search "query"	Search agent registry
agenteazy list	List all registered agents
agenteazy status	Show deployment status

Environment Variables

Command	Description
agenteazy env list	List configured env vars
agenteazy env set KEY VAL	Set an environment variable
agenteazy env remove KEY	Remove an environment variable

Infrastructure

Command	Description
agenteazy gateway deploy	Deploy gateway to Modal
agenteazy gateway status	Gateway health check
agenteazy registry deploy	Deploy registry to Modal
agenteazy registry start	Run registry locally
agenteazy stop <name>	Stop a deployed agent
agenteazy logs <name>	View agent logs

---

AgentLang — 10 Universal Verbs

Every agent speaks the same protocol. One HTTP POST, one envelope:

{"verb": "DO", "payload": {"data": {"input": "hello"}}}

Verb	Purpose	Status
DO	Execute a task	Working ✅
ASK	Query capabilities	Working ✅
FIND	Search the registry	Working ✅
PAY	Transfer credits between agents	Working ✅
SHARE	Pass context to an agent	Working ✅
REPORT	Get audit log and recent calls	Working ✅
STOP	Halt a running task	Working ✅
WATCH	Subscribe to events	Stub 🔜
TRUST	Establish authenticated session	Stub 🔜
LEARN	Ingest new knowledge	Stub 🔜

---

TollBooth — Agent Credits

Agents can charge credits per call. Developers earn 80%, platform keeps 20% for infrastructure.

agenteazy deploy github.com/you/ml-model --price 10

No billing code needed. Set a price and the gateway handles everything.

• Free agents: 0 credits per call (default)
• Paid agents: Set any price with --price
• Revenue split: 80% to agent developer, 20% platform fee
• Top up: agenteazy balance to check, credits added via signup

---

Architecture

Developer: agenteazy deploy <repo>
    |
    |-- 1. ANALYZE  -> Clone, detect Python, parse AST
    |-- 2. WRAP     -> Generate agent.json + FastAPI wrapper
    |-- 3. UPLOAD   -> Push to serverless volume
    |-- 4. REGISTER -> Add to public registry
    +-- 5. LIVE     -> Callable at gateway/agent/{name}/

Gateway (1 endpoint) ---- Registry (SQLite)
    |                          |
    +-- Agent Volume           +-- TollBooth (credits)
        /agents/repo-a/            /tollbooth/balance
        /agents/repo-b/            /tollbooth/deduct
        /agents/repo-c/            /tollbooth/earn

---

Supported Repo Types

Type	Supported
requirements.txt	✅
pyproject.toml	✅
setup.py	✅
Class-based agents	✅
src/ layout	✅
Environment variables	✅

---

Contributing

We welcome contributions:

1. Open an issue describing what you want to build or fix
2. We'll discuss the approach together
3. Clone the repo and create a feature branch
4. Submit a Pull Request referencing the issue

Questions? Open an issue or email [e-mail]

License

MIT — see LICENSE for details.