Codeligence Dev Agents¶

You have your copilot, but what about the agile team work outside the IDE? Dev Agents handle the nasty grind outside your IDE: docs, reviews, debugging, logs & delivery, so you peacefully focus on building.

Setup Wizard · Quick start · Use Cases

Core idea: one consistent, named AI teammate (e.g. “Betty Sharp”) embedded into Slack, GitHub, DevOps, Jira, Console & more - automating tedious, repeatable tasks around your codebase.

Build using the elegant Pydantic AI framework.

NEW: Execute Claude Code right from Slack!!

Why use Dev Agents¶

• Context and Integrations for Dev Teams. While it works similarly to Claude Code or Gemini CLI, Dev Agents provide you with pre-engineered, dev specific contexts, prompts, and integrations for reproducible, cost-efficient results and a quick start.
• Dev AI Avatar – one persona, many skills; shows up across your stack with a single voice.
• Open-source – free under AGPLv3; commercial license available for closed-source deployments, enterprise support and more use cases.
• LLM-flexible – works with major hosted or local models.
• Ready now – 3 production agents live; more available in the enterprise edition.
• Easily Customizable – Easily add more use cases or customizations by extending base classes and implementing protocols.
• Fast setup – guided onboarding at setup.dev-agents.ai; run locally or on your server.
• Built for teams – shines with 10 - 1000 dev teams and complex, evolving codebases where docs, compliance & handovers matter.

Quick start¶

1. Option: Run the Setup Wizard: setup.dev-agents.ai Generates your config and start instructions for local or server deployment.
2. Option: Use Docker:

# 1) Get example env and edit
wget -O .env https://raw.githubusercontent.com/codeligence/dev-agents/refs/heads/main/.env.example

# 2) Run Dev Agents in the docker container
docker run --rm -it --env-file=.env -v ./your-cloned-git-repo:/code codeligence/dev-agents

You need to mount your cloned Git repository into the container so that Dev Agents can work with the code.

1. Option: Clone and run:

# 1) Clone
git clone https://github.com/codeligence/dev-agents.git
cd dev-agents

# 2) Copy example env and edit
cp .env.example .env

# 3) Install and run Dev Agents
pip install -e .[all]
python -m entrypoints.main

Then interact in your tools (e.g. Slack): @DevAgents release notes for pull request 123 please

Use Cases¶

In the open source repo:

• Test Plan Generation – Generate a UI test plan from your pull request.
• Release Notes & Changelog – turns merged PRs into clear notes for products/libs.
• User Story Refining – improves stories with concrete, testable detail.
• Prod Log Root-Cause Analysis – surfaces likely cause, links to code, suggests fixes.

Some custom enterprise use cases:

• PR Review & Compliance Check – design patterns, conventions, risk flags.
• Support Reply Drafts – proposes informed responses from logs/context.
• Code Migration Assistant – highlights cross-repo testing considerations for framework/library jumps.

Who it’s for¶

• Engineering teams of 10 - 1000 devs on long-lived, multi-gen codebases
• Teams with documentation/compliance/support overhead
• CTOs who want to multiply output while protecting developer focus

Interfaces & Integrations¶

Interfaces

• Slack • Teams • AG-UI • MCP • CLI • A2A
• Add more easily

Integrations

• Git providers: GitHub / GitLab / Azure DevOps
• Issues/PM: Jira, GitHub Issues, GitLab Issues
• Observability: ELK / Loki / Files (others via MCP/tools)
• Models: All major providers and local LLMs
• Add more easily

Community & Support¶

• 📚 Documentation: Comprehensive guides and API reference
• 🐛 Issues: GitHub Issues
• 💬 Discussions: GitHub Discussions
• 💬 Discord: Join Discord Server
• 🔒 Security: Security Policy

Next Steps¶