TrueFoundry

TrueFoundry now integrates with @straikerai .

Teams routing LLM and agent traffic through the TrueFoundry AI Gateway can now add Straiker as a guardrail that checks every prompt on the way in and every model response on the way out: prompt injection, manipulated agents, risky tool calls, and multi-step attacks that don’t fit inside a single request, all enforced at the gateway, with no adapter to build and no changes to application code.

Security teams set the controls per agent in the Straiker console. Straiker returns an allow-or-block verdict in under 300 ms, and the gateway enforces it. Start in Audit mode to see what gets flagged, flip to Enforce when it looks right, and every check shows up in TrueFoundry’s request traces. Combined with TrueFoundry’s unified access to 2,500+ LLMs, rate limiting, cost tracking, and governance, platform teams keep one governed gateway and application teams ship agents without rebuilding security in each one.

One gateway. Every request, protected.

Thanks to the @straikerai team for the collaboration!

Something big is coming on 10 November 👀

On Tuesday, we start a new chapter for TrueFoundry, and it’s one we’ve been looking forward to for a long time. If you look after AI at your company, this one might be interesting.

Our TrueForge challenge at Stranger Builds, NERDCONF's Halloween hackathon: build an agent that scores itself, rewrites its own skills, and beats run 1 on a later run.

USD 10K in prizes and credits. Sat 31 Oct in person or online, then 7 days to ship.

Registration link below!

We're running a TrueForge track at Stranger builds: build a self-improving agent 🔁
an agent that gets better every run beats one that's perfect on day one. show us yours.
See you Saturday!