sekera-radim/impri

Overview

Impri is a self-hostable tool that adds a human-in-the-loop approval step to AI agent workflows. Agents submit proposed actions to an inbox, where a human can approve or reject them before execution. It supports MCP and REST interfaces and includes watcher functionality for monitoring proposals.

Best for

Best for
Teams needing a lightweight, self-hosted approval layer for AI agents

Use cases

• Adding human oversight to autonomous agent workflows
• Approving or rejecting agent-generated actions before execution
• Monitoring agent proposals with watchers

Notes

Impri is a self-hostable tool that adds a human-in-the-loop approval step to AI agent workflows. Agents submit proposed actions to an inbox, where a human can approve or reject them before execution. It supports MCP and REST interfaces and includes watcher functionality for monitoring proposals.

0 stars on GitHub. Last updated 2026-07-13. Licensed MIT.

Use cases

• Adding human oversight to autonomous agent workflows
• Approving or rejecting agent-generated actions before execution
• Monitoring agent proposals with watchers

Pros

• Self-hostable with MIT license for full control
• Provides clear human-in-the-loop governance for agent actions
• Supports both MCP and REST interfaces for integration

Cons

• No stars yet, indicating a new or unproven project
• Requires manual approval step, slowing fully autonomous processes
• Limited community support and documentation due to early stage

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.