Open-source AI Security Operations Center: alert fusion, LLM-agent triage, MITRE ATT&CK investigation, and a replayable decision ledger for every agent step. Self-hostable, runs with no API keys, MIT licensed. Ships an MCP server for Claude, Cursor and Continue.
Python agentic-aiai-securityalert-triageblue-teamcybersecuritydetection-engineeringincident-responselanggraphllm-agentsmcpmitre-attackmodel-context-protocolpurple-teamsecurity-operationsself-hostedsiemsoarsocthreat-detectionthreat-intelligence
chore(8.1): re-export the spec after the MFA policy docstring changed
Beenu Arora13c9a3e4d9e7
diff --git a/docs/openapi.yaml b/docs/openapi.yamlindex 7e19e62f5..39313cb64 100644--- a/docs/openapi.yaml+++ b/docs/openapi.yaml@@ -434,15 +434,26 @@ paths:`require_for_roles` names roles that must hold a factor even when the- tenant-wide switch is off. Every name is checked against+ tenant-wide switch is off. Every name is checked against the role- `ROLE_PERMISSIONS` before it is stored: a typo would otherwise be+ vocabulary before it is stored: a typo would otherwise be accepted,- accepted, match nobody''s role, and leave an administrator looking at+ match nobody''s role, and leave an administrator looking at a policy- a policy page that says a role is enforced while it is not — the+ page that says a role is enforced while it is not — the quiet- quiet direction, and the one worth refusing loudly.'+ direction, and the one worth refusing loudly.+++ The vocabulary comes from `role_grants.known_roles()` rather than from++ `ROLE_PERMISSIONS` directly. This route decides nothing about++ permissions and must not be able to: a route that reads the static map++ is how a second permission model gets rebuilt, which++ `scripts/check_one_permission_model.py` exists to prevent.'operationId: put_policy_api_v1_auth_mfa_policy_putrequestBody:content: