Skip to main content
POST
/
alignment
/
agent
/
{agent_id}
/
explain
Explain policy evaluation for the alignment card
curl --request POST \
  --url https://api.mnemom.ai/v1/alignment/agent/{agent_id}/explain \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "turn_id": "<string>",
  "request_snippet": {
    "tool_name": "<string>",
    "tool_args": {}
  },
  "enrich": true
}
'
{
  "ok": true,
  "reasoning": "<string>",
  "suggested_remediations": [
    {
      "index": 123,
      "remediation": "<string>",
      "method": "<string>",
      "url": "<string>"
    }
  ],
  "enriched": true,
  "trace": {}
}

Authorizations

Authorization
string
header
required

Supabase JWT token in Authorization: Bearer header

Path Parameters

agent_id
string
required

Agent identifier (e.g. smolt-abc123)

Body

application/json
turn_id
string

Optional — bind to a specific gateway turn.

request_snippet
object

Optional — hypothetical tool call to explain.

enrich
boolean

When true, runs an LLM pass over the structured remediations. Off by default.

Response

Policy evaluation trace + remediations.

ok
boolean
required
resource
enum<string>
required
Available options:
alignment
reasoning
string
required

Care-framed prose summary.

suggested_remediations
object[]
required
enriched
boolean
required

True when the LLM enrichment ran.

trace
object

EvaluationResult from @mnemom/policy-engine. Null for protection in V1.