Safe House

Retro-correct past Protection screening verdicts against the current card

Re-buckets the agent's past NON-clear Safe House screening decisions (sh_evaluations) against the agent's CURRENT composed Protection card thresholds and durably flips any that are no longer violations. Deterministic threshold re-bucketing, not a re-run of the L1/L2 detector models. Mirrors POST /v1/agents/{agent_id}/reverify/aip (tags/checkpoints.ts) structurally (MNE-7172).

post/agents/{agent_id}/reverify/protection

Request

  • Base URL: https://api.mnemom.ai/v1
  • URL: https://api.mnemom.ai/v1/agents/{agent_id}/reverify/protection
  • Auth: one of:
    • HTTP bearer
    • API key in header X-Mnemom-Api-Key

Path parameters

agent_idstring required

Agent identifier (e.g. smolt-abc123)

Query parameters

limitinteger
offsetinteger

Request body

changed_clausesstring[]

Optional list of Protection card clause paths that changed, recorded on the audit trail only.

Response

Retro-correction batch results

agent_idstring
card_idstring
total_in_batchinteger
offsetinteger
limitinteger
re_checkedinteger
clearedinteger
still_flaggedinteger
decisionsobject[]
errorsstring[] nullable

Changes

No recorded changes to this operation across all 1 revision of this API.