CalibratedDecisions.

Repo · Safety & guardrails

pi-reply-guard

Pi extension that checks each agent reply against installed skills you choose: a classifier — Jev by default (openrouter/typesafe/jev-1.13) — returns…

Open github.com ↗

From the repository

Checks each agent reply against a list of skills. A classifier model determines if the reply adheres to the skill on a scale of 0-1. If a reply does not adhere to the skill, the agent is prompted to edit their reply to adhere to the skill.

pipi-extensionpi-packagereply-guardskills

How builders describe it

Pi extension that checks each agent reply against installed skills you choose: a classifier — Jev by default (openrouter/typesafe/jev-1.13) — returns the probability the reply adheres to each skill, and replies below your threshold are sent back for a rewrite (up to maxRewrites).

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