Unverified50% confidenceBenchmarkExact time
On OpenClaw recruitment tasks across five models, this harness-layer guard reduced the violation rate from 65.28% to 6.67%.
1
Sources
50%
Confidence
Long-term
Relevance
10/27/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe same model (with identical weights) scores 62% in one agent harness, but only 33% in a different harness, a gap of nearly 30 percentage points.69% similarUnverifiedIn experiments, CarryOn reduced the safety violation rate from 52.6% to 6.1%.67% similarUnverifiedResearch indicates that an attacker needs to modify as little as 0.01% of training data to potentially implant backdoor behaviors in a model that are undetectable by standard evaluations.66% similarUnverifiedFor a task that sequentially calls 5 APIs, even if the success rate of each step is 95%, the overall success rate drops to approximately 77%.66% similarUnverifiedIn tests across five LLMs and three Agent harnesses, with newly added Skills capped at only 5% of each original pool, the policy violation rate rose from 35.02% to 58.45%, an increase of 23.43 percentage points.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1102070API
curl https://kongchang.com/api/v1/knowledge/claims/1102070MCP
get_claim(id=1102070)