Read scoped Prompt configuration and built-in defaults

GET
/v1/scopes/{scope_id}/prompts/{prompt_key}

Read one saved Prompt head and the current Runtime's built-in instructions without creating a revision or calling inference. Auto selects the built-in instructions even when an Auto revision exists. Disabled built-in operations remain readable; effective and builtin are null for externally managed components. Status describes availability, not whether returning the configured text executes it. A saved head additionally requires its Artifact read permission. artifact_etag is the condition for replacing that Artifact; it is not an ETag for this combined view.

Authorization

BearerAuth
AuthorizationBearer <token>

Bearer credential resolved to an opaque authenticated Principal by the Server deployment.

In: header

Path Parameters

scope_id*string
Match.*\S.*
Length1 <= length <= 256
prompt_key*string

Value in

  • "memory.extract"
  • "memory.rerank"
  • "experience.incubate"
  • "experience.generate"
  • "skill.generate"
  • "handoff.generate"
  • "profile.generate"

Response Body

application/json

application/json

application/json

application/json

application/json

application/json

application/json

curl -X GET "https://example.com/v1/scopes/string/prompts/memory.extract"
{  "scope_id": "string",  "prompt_key": "memory.extract",  "status": "supported",  "reason": "operation_disabled",  "mode": "auto",  "artifact": {    "family": "string",    "artifact_id": "string",    "revision": 1  },  "artifact_etag": "string",  "effective": {    "instructions": "string",    "demonstrations": [      {        "input": null,        "expected_output": null      }    ]  },  "builtin": {    "version": "string",    "profile": "coding",    "instructions": "string"  }}