@Bluebear_sec

Bluebear Security - Safe to imagine

 New York, NY
bluebear.io
Joined June 2026
  • Tweets4
  • Following39
  • Followers6
2Media
The majority of published coding agent security issues from June 2025 sit within the code around the model, and many don't require a model at all. We compiled our research and built a Claude skill to help you build secure harnesses, check it out here 👉 bluebear.io/blog/what-coding…

Lessons from a year of agent CVEs · Bluebear

We read every published security issue in AI coding agents from the past year. Most sit in the code around the model, and a quarter need no model at all.

bluebear.io
1
1
1
2
45
Ready to turn on YOLO mode for your AI coding agents? Come meet us at AI DevCon NYC to see how Bluebear secures agents at runtime & gives development teams confidence to trust agents with meaningful work. 👉 aidevcon.io/ @ainativedev
2
2
59
Bluebear is proud to integrate with @claudeai to bring runtime guardrails and user intent mapping to your Claude Enterprise rollout. Safeguard runtime. Set your ambitions free. Learn more about Bluebear's integration with Claude's Compliance API: lnkd.in/eARvRupE
2
2
114
Would your agent respect a block? One customer's agent hit a block at the MCP gateway, switched to Bash, and successfully submitted the change. This is why MCP-layer controls aren't enough, and why policy belongs at runtime. Read more: bluebear.io/blog/the-tool-bl…

The tool blocked it, but the agent kept going. · Bluebear

An MCP tool denied a production change, so the control looked like it worked. The agent switched to Bash and made the same change anyway. A block on one tool does not prove the action is blocked.

bluebear.io
2
1
3
53
Load more