Skip to main content
POST
Start People Intelligence

Authorizations

x-api-key
string
header
required

Your Sixtyfour API key.

Headers

Idempotency-Key
string | null
  • Scoped to organization + operation.
  • Same key + body: same job, no new charge, Idempotent-Replayed: true.
  • Different body: 409 idempotency_key_reused.
  • Uploads must match filename + contents.
  • Omit for no replay protection.
  • JSON replay lasts through the workflow retention window; CSV replay lasts 24 hours.
Required string length: 1 - 255

Body

application/json
lead_info
Lead Info · object
required

The person to look up: any combination of name, email, company, linkedin and other known details.

struct
Struct · object | null

Fields to return: output field name to a description of what to find. An image field returns a https://api.sixtyfour.ai/v1/media/... URL you can embed. Required except on Atlas runs (xhigh, or scout with graph).

research_plan
string | null

A plan, in plain language, that guides the research. Default: none.

Maximum string length: 100000
tier
enum<string> | null

Research depth and price. See Credits & Pricing. Default: low.

Available options:
low,
micro,
medium,
high,
xhigh,
scout
webhook_url
string<uri> | null

HTTPS URL that receives the result when the job completes. Default: none.

Required string length: 1 - 2048
field_confidence
boolean | null

If true, returns each requested field's confidence (0-100), justification and source URLs, nested like the request. Default: false.

graph
boolean | null

scout only. If true, runs Scout as an Atlas investigation, billed per unit of research. It enables the Atlas run options (max_credits, max_minutes, relentless, investigation_id, investigation_name, include_workspace), makes struct optional, and the async response returns the investigation. Requires Scout in Atlas for your organization. If false or omitted, Scout runs as usual.

include_workspace
boolean | null

Atlas runs only. If true, includes the investigation's workspace files in the result; they are read from the investigation, so they include later runs' changes. Default: true.

investigation_id
string | null

Atlas runs only: continues this investigation, reusing its graph and workspace, instead of creating one. Default: creates one.

Maximum string length: 200
investigation_name
string | null

Atlas runs only: the title of a new investigation. Ignored with investigation_id. Default: taken from lead_info.

Maximum string length: 1000
private
boolean
default:false

If true, shows the run's result, status and usage log only to you (with an API key, to the key's creator). Your team must allow private runs. Atlas runs need the async endpoint and a new investigation.

connections
string[] | null

Your team's connected APIs the run's agents may call, by name. Leave it out to allow all of them; an empty list allows none. Taken by xhigh, by scout with graph, and by high; refused on other tiers.

Maximum array length: 10
Maximum string length: 63
relentless
boolean
default:false

Atlas runs with struct only. If true, researches until the budget runs out. If false, a run may stop once every requested field is answered, and is charged for the research it used (at least one unit). Default: false.

agent_trace
boolean
default:false

Atlas runs only. If true, the investigation's Agents tab in Atlas shows a short, readable summary of each agent's reasoning and actions as it works. The trace is for people reading along: it does not change the research or the results. Leave it off when you only read results through the API. Default: false.

generate_report
boolean
default:false

Atlas runs only. If true, starts the investigation's PDF intelligence report when research completes. You can also generate one at any time with POST /investigations/{investigation_id}/report or from the Report tab in Atlas. Default: false.

max_credits
integer | null

Atlas runs only: research budget in credits, from one to 10 units of the tier price. Default: one unit. The whole budget is used and charged unless the run stops early (see relentless and max_minutes).

max_minutes
integer | null

Atlas runs only: the most minutes of research time; waiting to retry does not count. When reached, the run writes up its findings, which can take a few minutes, and is charged only for the research it used. Default: no limit.

Required range: 10 <= x <= 1440
skills
string[] | null

Skills the run's agents follow, by name: team/<name> for one of your team's skills, sixtyfour/<name> for one of ours. Up to 25,000 tokens in total. Taken by xhigh, by scout with graph, and by high; refused on other tiers.

Required string length: 1 - 74
exclusive_run
boolean
default:false

Atlas runs on an existing investigation only. If true, returns 409 while that investigation has a run starting or running, instead of starting and charging a second one. Default: false.

Response

Successful Response

job_id
string
required

Job identifier, normally prefixed with job_. Store it exactly as returned. GET /v1/jobs/{job_id} also accepts IDs returned by unversioned endpoints.

status
enum<string>
required

queued: accepted and waiting to start. running: work is in progress. succeeded: work finished successfully. failed: work ended with an error. canceled: work was canceled. The last three states are terminal; stop polling.

Available options:
queued,
running,
succeeded,
failed,
canceled
created_at
string<date-time>
required

UTC timestamp when the job was created, in ISO 8601 format.

next_poll_after
integer | null

Seconds before polling again while queued or running; omitted for terminal states.

progress
Progress · object | null

Operation-specific progress while queued or running; omitted when unavailable.

result
any

Successful output when available; bulk jobs may return download links.

error
JobError · object | null

Failure details when available; omitted otherwise.