Agent Network Understanding - Structured Outputs
Iam trying to understand if agent networks can create
- structured otuput
- use output processors in the sub agents adn aceess subagents messages response ( which is what output processors use )
Basically my goal is to acces on the network the agents result that was processeded through a ouput processor ??
is there a example explaintion for it somewhere ?
iam letting the subaget create some json
though if using gemini it likes to create
json with backticks... and i need to strip this away with otuput processor 😄
```
7 Replies
📝 Created GitHub issue: https://github.com/mastra-ai/mastra/issues/10750
🔍 If you're experiencing an error, please provide a minimal reproducible example whenever possible to help us resolve it quickly.
🙏 Thank you for helping us improve Mastra!
AgentNetwork does not return structured output or output processors.
The network turns the agent into a Router, if the subagent has processors and all that they will be run when the agent is executed
We will make some changes to network to be able to be more configurable in the new year
ohh okay ... thats a nerf compared to workflows then. not really fitting my usecase then
The code for agent network is more valuable than the actual implementation cause it is just a workflow!
GitHub
mastra/packages/core/src/loop/network/index.ts at main · mastra-ai...
The TypeScript AI agent framework. ⚡ Assistants, RAG, observability. Supports any LLM: GPT-4, Claude, Gemini, Llama. - mastra-ai/mastra
You can see what we did here and make your own!
isee thanks