Preview an auto-invite rule
Counts how many of the job's existing finished reports a draft rule would match, would not match, or cannot judge, split by report version, so you can choose a bar before saving it. Also counts newMatches, the existing reports applyToExisting would send through the rule, and for a draft in invite mode the applyQuote: the most credits that could spend. Saves nothing and changes nothing. Scores from different report versions are on the same 0 to 10 scale but are not calibrated alike, which is why the counts are split.
Required scope: auto_invite:read (auto_invite:write also works).
Errors
- 400:
invalid_content_type,invalid_json,request_body_too_large,unknown_field - 401:
authentication_required,invalid_api_key - 403:
api_access_disabled,insufficient_scope - 404:
job_not_found - 422:
invalid_parameter - 429:
rate_limit_exceeded - 500:
api_error - 503:
maintenance_mode,service_unavailable
ApiKeyAuthorizationBearer <probb_live_<id>_<secret>>id*stringThe job's ID.
1 <= length <= 200application/json- body
criteria*The bar a finished report must clear to be queued. All conditions must hold.
mode?stringThe mode you intend to save, for applyQuote only. Omit to use the rule's current mode (queue for a job without a rule).
"queue""invite"inviteCap?integerThe cap you intend to save, for applyQuote only. Omit to use the rule's current cap.
1 <= value <= 500Success.
application/json- response
object*stringtotal*integerExisting finished reports on the job that were read (reports held as role mismatches are left out).
-9007199254740991 <= value <= 9007199254740991wouldMatch*integer-9007199254740991 <= value <= 9007199254740991wouldNotMatch*integer-9007199254740991 <= value <= 9007199254740991unknown*integerReports the rule cannot judge, because their version has no requirements check.
-9007199254740991 <= value <= 9007199254740991withRequirementsCheck*integer-9007199254740991 <= value <= 9007199254740991byVersion*array<>truncated*booleanThe job has more finished reports than the preview reads.
newMatches*integerExisting reports applyToExisting would send through the rule: ones it would queue or hold now, with no waiting or settled evaluation of this rule, nothing the bulk invite would skip (an interview, no email, a role mismatch) and no hiring decision on the candidate. At most 1,000 reports are read.
-9007199254740991 <= value <= 9007199254740991newUnknown*integerOf newMatches, the reports whose must-haves cannot be checked. With whenUnknown: queue they go to a person in any mode and are never invited automatically, so applyQuote leaves them out.
-9007199254740991 <= value <= 9007199254740991applyQuote*|What applyToExisting could spend. Present only when the draft mode is invite and newMatches is above zero. Send it back as maxCredits on PUT.
# Set PROBB_URL and PROBB_API_KEY. Replace any *_VALUE placeholder and any example ID (such as job_example) with a real one.curl --fail-with-body -X POST "$PROBB_URL/api/v1/jobs/ID_VALUE/auto-invite/preview" \ -H "Authorization: Bearer $PROBB_API_KEY" \ -H 'Content-Type: application/json' \ --data-binary '{ "criteria": { "minFitScore": 7.5 }}'{ "object": "auto_invite_preview", "total": -9007199254740991, "wouldMatch": -9007199254740991, "wouldNotMatch": -9007199254740991, "unknown": -9007199254740991, "withRequirementsCheck": -9007199254740991, "byVersion": [ { "versionId": "string", "label": "string", "hasRequirementsCheck": true, "total": -9007199254740991, "wouldMatch": -9007199254740991, "wouldNotMatch": -9007199254740991, "unknown": -9007199254740991 } ], "truncated": true, "newMatches": -9007199254740991, "newUnknown": -9007199254740991, "applyQuote": { "billingMode": "CREDITS", "creditsPerInvite": 0, "affordable": true, "capRoom": -9007199254740991, "invites": -9007199254740991, "credits": 0 }}