#9696th of 142 in Agent frameworks and SDKs
sprix-sage-router
Python router that picks continue, collaborate or hand off for in-flight A2A tasks
- Stars
- 3.6k
- License
- MIT
- Last commit
- Aug 2026
- Last release
- Aug 2026
- Language
- Python
359 on 19 Aug 20263,562 on 11 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
SAGE is a Python reference implementation of a routing layer between A2A agent discovery and task execution. Given a partially completed task DAG, it scores three routes (SELF, COLLABORATE, HANDOFF) under budget and deadline limits, using per-requirement trust and artifact reuse, and returns an auditable decision. It is a research preview, and its benchmarks use synthetic evaluators written with the project.
Who it is for: Researchers and engineers building multi-agent A2A task routing
Strengths
- No runtime dependencies; Python 3.10+ and `python demo.py` to try it
- Decision traces list excluded agents and alternatives, so routing choices are auditable
- Learned trust state can be exported as versioned JSON snapshots
- Deterministic seeded benchmarks with JSON output; README states their synthetic limits
Weaknesses
- Does not send tasks, authenticate endpoints or verify signatures; client handles A2A transport
- Benchmarks use synthetic evaluators only; no real A2A endpoint evaluation yet
- Real A2A adapters and signed Agent Card ingestion are unchecked on the roadmap
- Labelled early-stage research preview with no production SLA
What it needs
- no GPU
What changed
From the release notes. No flag means the note does not say so, not that the upgrade is safe.
Also in Agent frameworks and SDKs
See all 142| Rank | Project | Score |
|---|---|---|
| 1 | ppt-masterAgent skill that turns documents into natively editable PowerPoint decks | 96 out of 100 |
| 2 | openai-agents-pythonPython SDK for multi-agent workflows with handoffs, guardrails and tracing | 92 out of 100 |
| 3 | mastraTypeScript framework for building AI agents, workflows and MCP servers | 92 out of 100 |
| #4 | browser-usePython library that lets an LLM agent control a web browser | 90 out of 100 |
| #5 | copilotkitReact and Angular SDK for adding agent chat and generative UI to apps | 90 out of 100 |