# Boundrix (boundrix.io)

Authorizes each tool and API call an autonomous AI agent makes against a task-scoped policy, allowing, denying or escalating it before execution.

- Atlas card: https://inite.ai/en/atlas/boundrix.io
- Site: https://boundrix.io
- For: Engineering and security teams running autonomous AI coding agents against GitHub repositories
- Categories: dev
- Use not measured/100 · alive 100/100

## Used for

- scope coding agent actions to assigned repositories and branches
- block destructive actions and credential exfiltration by agents
- revoke an agent's access mid-task without ending its session

## Price

Not read yet.

## For developers

None found.
- Read 2026-10-28

## Your data

- Certifications stated: none
- Training on your data: no (https://boundrix.io/security)
- Read 2026-10-28

---

Source: INITE Atlas, https://inite.ai/en/atlas/boundrix.io. JSON: https://inite.ai/api/atlas/boundrix.io. MCP: https://inite.ai/api/mcp (atlas_lookup).