Skip to main content
GET
Get a cell's enrichment run detail

Authorizations

Authorization
string
header
required

Bearer authentication header of the form Bearer <token>, where <token> is your auth token.

Path Parameters

table_id
string<uuid>
required
record_id
string<uuid>
required
column_id
string<uuid>
required

Response

Successful Response

One cell's most recent enrichment execution, for inspection.

raw_response is the model's STRUCTURED payload (the parsed fields, pre-coercion) — never the provider transcript; summaries and error text stay sanitized and length-capped, so nothing here needs masking after the fact.

action_key
string
required
cached_tokens
integer | null
required

Prompt tokens served from the provider's cache; null when unreported.

completion_tokens
integer | null
required

Provider-reported completion tokens; null when unreported.

confidence
enum<string> | null
required

The agent's own read on the answer; null for non-agent actions.

Available options:
high,
medium,
low
created_at
string<date-time>
required
credits_charged
number
required

What executing this cell cost, by the page charge's rule.

details_pruned
boolean
required

Whether the run is past detail retention. When true, empty trail fields (steps, reasoning, sources) mean the detail aged out, not that the run never had one.

duration_ms
integer
required
error
string | null
required
error_class
enum<string> | null
required

Machine-readable classification of a failed enrichment cell.

The class decides everything downstream: whether the executor retries (transient classes) or gives up (terminal ones), what the cell tooltip names, and whether the failure is free — errors never charge, but only a classified error can prove which kind it was.

Available options:
timeout,
rate_limited,
provider_error,
network,
schema_violation,
refused,
invalid_input,
no_data,
enrichment_failed,
unexpected
id
string<uuid>
required
model_key
string | null
required
outcome
enum<string>
required

How one cell's enrichment execution ended.

SKIPPED is a first-class third outcome: the action was deliberately not run (a required input was empty), no work was done, and nothing is charged.

Available options:
succeeded,
failed,
skipped
prompt_text
string
required

The resolved per-row prompt the action received.

prompt_tokens
integer | null
required

Provider-reported prompt tokens; null when unreported.

raw_response
Raw Response · object | null
required

The model's structured payload exactly as returned, before coercion; null when the action produced none.

reasoning
string | null
required
reasoning_tokens
integer | null
required

Hidden reasoning tokens billed as output; null when unreported.

run_id
string<uuid> | null
required

The table run that executed the cell.

source_attempts
EnrichmentRunSourceAttemptSchema · object[]
required

A curated enrichment's per-attempt trail; empty elsewhere.

sources
EnrichmentRunSourceSchema · object[]
required

The pages the agent successfully read, deduplicated.

steps
EnrichmentRunStepSchema · object[]
required
total_tokens
integer | null
required

Provider-reported total tokens; null when unreported.

web_read_credits
integer | null
required

Web-read credits the run's page fetches billed, summed across steps; null for cells without web tools or rows predating the telemetry.