NetworkManager Works to Enforce AI Policy by Tricking AI Agents to Add a Canary
22 days ago
- NetworkManager adopted an AI coding policy requiring authors to take full responsibility and explain all code they submit.
- To enforce the policy, the project introduced a canary word 'biblioklept' in AGENTS.md to trick AI agents into adding it to commit messages.
- Patches containing the canary word are automatically detected and discarded to reduce maintainer burden.
- Developer Josephine Pfeiffer implemented this approach using AGENTS.md / CLAUDE.md documentation.