Agent classes / Security review agent
Security review agent Infrastructure Kit
Reads a change or a trace for secrets, overly broad permissions, and untrusted text copied into a tool prompt. It does not produce an attack.
Prompt-injection / jailbreak firewall
Depot Pick: Lakera Guard
Programmable LLM input/output guardrails
Depot Pick: NVIDIA NeMo Guardrails
LLM firewall
Depot Pick: Arthur AI
Red-team evals
Depot Pick: Promptfoo
Agent secrets & non-human credentials
Depot Pick: Infisical
Secret rotation
Depot Pick: Doppler
PR review agents
Depot Pick: Anthropic