squadron
mcp
Warn
Health Warn
- License — License: MIT
- Description — Repository has a description
- Active repo — Last push 0 days ago
- Low visibility — Only 6 GitHub stars
Code Pass
- Code scan — Scanned 12 files during light audit, no dangerous patterns found
Permissions Pass
- Permissions — No dangerous permissions requested
No AI report is available for this listing yet.
Declarative multi-agent AI workflows in HCL. Single Go binary. Native MCP. Plugins in Go and Python.
README.md
Squadron
AI agent workflows as configuration, not code.
Squadron is a framework for defining and running multi-agent missions in HCL. You declare agents, tools, and task graphs in config files — Squadron handles orchestration, state management, branching, and recovery.
Install
curl -fsSL https://raw.githubusercontent.com/mlund01/squadron/main/install.sh | bash
Or download from GitHub Releases. See the installation docs for more options.
Quick Start
# Generate a sample workflow
squadron helloworld && cd helloworld
# Set your API key
squadron vars set anthropic_api_key sk-ant-...
# Launch the command center
squadron serve -w
Example Mission
mission "data_pipeline" {
commander = models.anthropic.claude_sonnet_4
agents = [agents.assistant]
task "fetch" {
objective = "Fetch data from the API"
output {
field "records" { type = "string"; required = true }
}
}
task "process" {
depends_on = [tasks.fetch]
objective = "Analyze the data and produce a summary"
}
task "review" {
depends_on = [tasks.process]
objective = "Review the analysis"
router {
route {
target = tasks.escalate
condition = "Issues were found that need human review"
}
}
}
task "escalate" {
objective = "Flag issues and prepare an escalation report"
}
}
squadron mission data_pipeline -c ./config
Docker
docker run -v ./config:/config -v squadron-data:/data/squadron -p 8080:8080 \
ghcr.io/mlund01/squadron engage
Alpine (default) and Debian images are published to ghcr.io/mlund01/squadron on every release. See the Docker docs for details.
Key Features
- HCL configuration — agents, tools, and missions defined as config files
- Mission orchestration — task dependencies, LLM-driven routing, fan-out, cross-mission routing
- Persistence and resume — resume interrupted missions from where they failed
- Command center — web UI for running missions, DAG visualization, live monitoring
- Multi-provider — Anthropic, OpenAI, Google Gemini
- Plugin system — extend with plugins for browser automation, APIs, and custom integrations
Providers
- Anthropic (Claude)
- OpenAI (GPT-4)
- Google Gemini
Plugins
Build custom plugins with the squadron-sdk.
License
MIT
Reviews (0)
Sign in to leave a review.
Leave a reviewNo results found