Vibeleaderboard
← All Intel
Intel / repo

AI agent follows instructions you never approved

Source
github.com
Author
paraphern
Date
Why it matters

tool descriptions are instructions agents obey, and this audit shows they change between releases without notice. Hash pinning and diffing contracts in CI catches silent removals and injected instructions before agents act on them.

Terms in this piece · Glossary
  • MCP — The Model Context Protocol — an open standard that lets any AI assistant plug into any tool or data source without custom integration code.
  • AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
Recommended reads
Comments

Checking sign-in…

Loading comments…