CalibratedDecisions.

Repo · Safety & guardrails

Approval Judge Bridge

OpenAI-compatible /v1/chat/completions proxy that answers an agent host's approval-guardian call with APPROVE, DENY or ESCALATE.

Open github.com ↗

From the repository

OpenAI-compatible judge endpoint for agent approval gates: typed judgements (Jev), any OpenAI-compatible model, or a rule file — fail-closed, calibrated, with a replay battery

agentai-safetyapprovalsguardrailshermesjevtypesafe

How builders describe it

OpenAI-compatible /v1/chat/completions proxy that answers an agent host's approval-guardian call with APPROVE, DENY or ESCALATE. Jev answers one Choice question with a calibrated distribution; missing keys, timeouts and malformed answers all fail closed to ESCALATE.

The decision Jev makes

Allow, flag or escalate a message, tool call or action.

Where it fits

Screening messages, tool calls and outputs before they cause trouble. Jev gives a fast, calibrated verdict that can sit in front of every request or agent action. All 229 safety & guardrails projects →

Related projects