PermitQueue: human approval gateway for agent tool calls

High-risk agent actions wait for a human to say yes

Problem

Coding agents increasingly want to edit files, deploy, and call external APIs directly, but many teams only let them run low-risk operations. Today's defenses are prompt-level constraints or after-the-fact audits; there is no unified gate that pauses execution and asks a human to approve, especially across MCP tools and local commands.

Solution

A lightweight approval gateway: every agent-initiated tool call passes through a rules engine, and calls that hit high-risk rules pause and generate an approval card pushed to web or Slack. Execution happens only after human approval, with a full audit log.

Users

Developers using coding agents locally or in CI; small engineering teams needing compliance trails.

MVP

Configurable high-risk rules matched by tool, command or path; call pausing with approval card push; one-click approve/deny in web and Slack; execution result returned to the agent; per-session audit log export.

Revenue

One-time purchase for the local version; monthly team subscription with approval record retention and multi-approver support.

Why now

MCP exploded the number of tools agents can call, yet human-in-the-loop approval is scattered across IDE plugins; there is no neutral, self-hostable gateway.