Skip to main content
GET
Retrieve Prompt Run

Authorizations

X-Lightsage-Api-Key
string
header
required

Path Parameters

run_id
string
required

Response

Successful Response

id
string
required
prompt_id
string
required
run_date
string
required
prompt
V2PromptRunPrompt · object | null
platform
V2PromptRunPlatform · object | null
target_kind
enum<string> | null
Available options:
answer_engine,
coding_agent
harness
string | null

Coding-agent harness for agent runs; otherwise the answer platform identifier.

model_id
string | null

Public model ID. Internal hosting providers are never returned.

response_preview
string | null
is_brand_mentioned
boolean | null
citation_count
integer
default:0
tool_call_count
integer | null

Coding-agent tool invocations, counting a call and its result once. Matches the length of tool_calls on the detail response; null for answer-engine runs.

search_grounding
enum<string> | null

Web access mode requested for this run. Null for runs recorded before web access was tracked.

Available options:
disabled,
enabled,
default
web_retrieval_observed
boolean | null

Whether the coding agent was observed searching or fetching from the web. Null when not measured, including answer-engine runs.

search_observed
boolean | null
grounding_qualified
boolean | null

Whether the run satisfied the requested web access mode. Null means the default mode was used or isolation could not be proven.

grounding_qualification_reason
string | null
created_at
string | null
response
string | null
mentions
V2PromptRunMention · object[]
citations
V2PromptRunCitation · object[]
tool_calls
Tool Calls · object[]

Sanitized coding-agent tool invocations captured for this run, one entry per invocation with its result merged in.

conversation_turns
Conversation Turns · object[]

Sanitized interleaved text and tool-call trace captured for this run. It interleaves the agent's text turns with tool_call turns, so its length is not a tool-call count; use tool_calls or tool_call_count.