Think of this as a black box recorder for your AI agents working with infrastructure. When your agents are provisioning servers, modifying configs, or orchestrating deployments, Evidra captures what they're doing and why they're doing it. You'd reach for this when you need audit trails, debugging context, or post-mortem analysis of agent actions in production environments. The flight recorder analogy is apt: it's not about preventing crashes, it's about understanding what happened when things go sideways. Useful if you're running autonomous or semi-autonomous agents that touch critical infrastructure and you need observability beyond basic logging.
EVIDRA_ENVIRONMENTEnvironment label (production, staging, development)
EVIDRA_EVIDENCE_DIRDirectory for evidence JSONL storage
EVIDRA_SIGNING_MODESigning mode: strict (default) or optional (ephemeral local mode)
EVIDRA_SIGNING_KEYsecretBase64-encoded Ed25519 private key (required for strict mode)
EVIDRA_SIGNING_KEY_PATHPath to PEM-encoded Ed25519 private key (required for strict mode if key not inline)
silenceper/mcp-k8s
azure/containerization-assist
io.github.evozim/aws-builder
reza-gholizade/k8s-mcp-server
flux159/mcp-server-kubernetes