This release adds 3 notable features for engineering teams evaluating rollout.
Published 10d
MCP Developer Tools
✓ No known CVEs patched
✓ No known CVEs patched in this version
Topics
agent-governance
ai-agents
go
guardrails-ai
llmops
llms
+8 more
mcp
observability
opentelemetry
python
reliability
self-hosted
sre
typescript
Summary
AI summaryUpdates @prashar32, examples, and adapters across a mixed release.
Full changelog
Changelog
- ab2ce718e6ac7d1e8cdd9957bb13cba8abf94472: docs(changelog): record the tool_calls audit read-back (#38) (#51) (@prashar32)
- aee4ea2411dc1b3de8b73031a41e4a7cb4155666: docs(examples): LangChain agent stopped at its budget (#56) (@prashar32)
- d1003e010c4b6bcc66e94a5f71584f7612b97b32: docs(examples): MCP gateway governs an agent's tool calls (#58) (@prashar32)
- 801123c91678eb9a6210d7415843d4c9f026310a: docs(examples): minimal "wrap your agent" governed loop (#54) (@prashar32)
- 42be464ad6af79e559d991197d4324d9d3eff132: feat(config): safe default budgets when none configured (#52) (@prashar32)
- 12aadf99b3065ac56f3320743d57a21c231eeaad: feat(pricing): config-updatable token→$ rates via RISKKERNEL_PRICING_FILE (#59) (@prashar32)
- ace8dad80c9dedae87d19f5b50c7e99dae9a9358: feat: surface tool call audit trail (#50) (@Sebastefanelli)
- 8b017b6f743cbdf6cdfdfd368bc045cec2d6ecde: fix(adapters): propagate budget halts out of the LangChain callback (#55) (@prashar32)
- bc0f587f452dedd3f782fc3b13e37d74372aa8ae: release: v0.2.0 (#60) (@prashar32)
- ad8ba0a7809cb1bf27a31d85d7a38e5b175bb3aa: test(adapters): lock in OpenAI Agents hook enforcement (#57) (@prashar32)
Self-hosted. Your keys. No telemetry. See SECURITY.md to verify.
Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
Share this release
About RiskKernel
All releases →Related context
Beta — feedback welcome: [email protected]