b3 Agent Security D
AI benchmark · Agentic
What is b3 Agent Security?
b3 (backbone breaker benchmark) measures how well backbone LLMs inside AI agents resist crowdsourced adversarial attacks such as prompt exfiltration, phishing link insertion and tool-call poisoning.
Security mostly
“we apply this framework to construct the b3 benchmark, a security benchmark based on 194,331 unique crowdsourced adversarial attacks.”
arxiv.org/pdf/2510.22620 ↗
Who built b3 Agent Security
Led by Lakera AI.
Paper: arxiv.org/pdf/2510.22620 ↗
Tier D
Long tail: little or no usage data yet.
How much stands behind a tier: usage, who backs it, a confirmed line from its own site or paper, and a clear read on X. High with three of the four, medium with two, low otherwise.
- no usage data
- in curated RL directories (1 pt)
How we got to tier D
- Adoption: no usage data. Tier D.
- Not yet read on X.
- In curated RL directories: 1 point, a floor of none.
- Confirmed: a line from its own site or paper places it.
A clear view on X moves a tier one step up or down; most entries don't move. How tiers work.
Where b3 Agent Security sits
Its domain on the map, live. Hover a logo for its name; click to open it. Open it on the full map
Next: browse all benchmarks. Entry from the RL Research daily scrape of public sources, 2026-10-30 snapshot.