» Tag
autonomous-agents
4 postsContext Bombs: Using AI Safety Guardrails to Halt Rogue Agents
Tracebit research shows context bombs hidden in canaries can trigger AI safety guardrails, cutting autonomous attacker success rates by roughly 90%.
mtime is not a claim: how one cp broke three monitoring systems
A cp without -p reset 54 file timestamps, fooling three separate monitoring systems. Why mtime should never be trusted as a freshness signal.
MoltProof Verifies Whether Autonomous Trading Agents Keep Their Word
MoltProof is a read-only verifier that proves whether on-chain autonomous trading agents actually followed the rules they publicly committed to, checkable by anyone.
CommitBrief — AI code reviews, right in your terminal
A provider-agnostic, local-first CLI that reviews your staged changes, a historic range, or a whole GitHub pull request. Zero telemetry, no server. Free and open source.
commitbrief.comChallenges of Running 9 Autonomous Agents in a Real Gym
Unexpected challenges and lessons learned from running 9 autonomous agents in a real gym.