Skip to content

AI Security

LLM security: prompt injection, jailbreak detection, guardrails, and adversarial evaluation.

Subscribe
← Releases
Review required
msaad00/agent-bom v0.98.1 Breaking risk
Auth Breaking upgrade

deploy, auth, trial, ui, readme

v0.95.0 (13d) UI, CLI, Deploy, Compliance, API
Review required
Guardrails v0.23.0 Breaking risk
Auth RBAC

Pydantic migration

No immediate action
h5i v0.2.2 Breaking risk

Env changes + cleanup + UI recenter

v0.7.8 (1mo) Browser folder upload removal
v3.17.0 (1mo) VG126 behavior change
Upgrade now
MikkoParkkola/nab v0.12.0 Breaking risk
Dependencies

CI, LinkedIn, Fingerprint, Task Engine, Python, Deps

v3.15.0 (1mo) AST/dataflow engine + VG406 improvements
v3.14.2 (1mo) VG964 false‑positive fix
v3.6.0 (1mo) SSRF false‑positive reduction
Config change
Guardrails v0.22.0 Breaking risk
Auth

Usage reporting, LangChain optional, IORails enhancements

v3.1.16 (2mo) Credential‑required connection strings
v3.1.4 (2mo) ADD matching, USER/HEALTHCHECK skipping, .env file restriction
v3.0.57 (2mo) SDK flag removals
v3.0.54 (2mo) Removed markdown library names
v3.0.26 (3mo) VG964 Next.js requirement

Beta — feedback welcome: [email protected]