Back to catalog / Agent workflows

Coordinate specialists into a coherent proposal

Assign domain questions to specialist agents and synthesize a consistent decision proposal.

Goal

Goal: Assign domain questions to specialist agents and synthesize a consistent decision proposal.

Define the deliverable and decision points; give each specialist a bounded role, source materials, and response format. [specialists]
Create specialists with make_agent(name, description, system, tools) and grant only role-relevant tools. [specialists]
Have the coordinator dispatch work, combine dependent findings, and preserve unresolved conflicts instead of inventing consensus. [specialists]

Check the proposal for each specialist’s evidence, dependencies, unresolved items, and tool boundaries.

If tasks overlap or conflict, clarify boundaries and preserve dissent; do not force artificial consensus.

Python 3.11+, Anthropic API key, Managed Agents beta SDK, and non-sensitive case materials that may be shared.

Bound each specialist’s data and tools; split confidential data only as authorized and prevent irrelevant cross-role exposure.

This is a proposal workflow; control parallel costs and require domain-expert sign-off for consequential decisions.

These values describe the upstream repository, not popularity or likes for an individual skill or playbook.

No metrics snapshot is available yet.

License: MIT