Framework for building, optimizing, and deploying AI agents as Effect services. Define an agent as a declarative blueprint. Run it against benchmarks.
Stars
2
Forks
0
Watchers
2
Open Issues
0
Overall repository health assessment
4.0.0-beta.434.0.0-beta.434.0.0-beta.43^1.6.1^0.1.14.0.0-beta.43^1.3.10^7.0.0-dev.20260401.14.0.0-beta.43^5.9.3^0.1.15^4.1.0^4.0.0-beta.4316
commits
Update README to reflect current state of blueprint-driven architecture
c25fc46View on GitHubRewrite program.md for blueprint workflow and add integration tests
8ca75e2View on GitHubWire --blueprint CLI flag and add serve http/mcp subcommands
64842fcView on GitHubImplement PlanAndExecute, WithVerifier, and FallbackModels orchestration strategies
4240ce1View on GitHubAdd deployment surface: HTTP API and MCP server
5dfe23fView on GitHubAdd runtime services: ToolFactory, BlueprintStore, AgentFactory
b2ed9f8View on GitHubAdd declarative agent schemas: ToolSpec, OrchestrationSpec, AgentBlueprint, BlueprintPatch
b9ac57aView on GitHubAdd 3 benchmark tasks and copy test scripts into container before verifier
be83564View on GitHubFix bench: run agent and verifier in the same per-task Docker container
5be92dcView on GitHubUpdate model references: gpt-5 → gpt-5.4, claude-sonnet-4-20250514 → claude-sonnet-4-6
2550869View on GitHubFix Docker execution: copyToContainer hang, output dir creation, examples use Docker
ce3e95bView on GitHubAdd README tailored to the effect-autoagent codebase
d187422View on GitHubAdd Effect v4 source as submodule at .references/effect-v4, pinned to effect@4.0.0-beta.43
40cc9afView on GitHub