Profile
Adversarial multi-perspective council MCP server. Multiple AI viewpoints debate before the agent commits to a decision.
Signals
Listed in the awesome-hermes-agent README
Sources: 2 / Surfaces: 1
What the upstream surface says
Short excerpt only, so you can decide whether to click out.
[](https://github.com/Ridwannurudeen/hermes-council/actions/workflows/ci.yml) [](https://github.com/Ridwannurudeen/hermes-council/actions/workflows/ci.yml)
Adversarial multi-perspective council MCP server for hermes-agent.
Five personas from distinct intellectual traditions debate questions through structured adversarial deliberation, producing calibrated verdicts with confidence scores, evidence links, and DPO preference pairs for RL training.
- multi-perspective-analysis — Full council deliberation workflow
- bayesian-synthesis — Bayesian reasoning and evidence synthesis
- adversarial-critique — Stress-testing and safety gating