Procedures manual

How to use ResumeGuard.

Each chapter is numbered and self-contained. Read top to bottom or jump to a section. All examples use the ResumeGuard API.

Chapter I

Getting started

Three steps separate you from your first verified result.

01Create an account at /signup. Trial credits are credited on confirmation.
02Issue an API key from the console (Settings → API Keys).
03POST a resume to /api/v1/document and inspect verification findings under GET /api/v1/result/:id.

curl -X POST https://api.resumeguard.ai/api/v1/document \
  -H "Authorization: Bearer $RG_API_KEY" \
  -F "resume=@candidate.pdf"

Chapter II

How ResumeGuard works

Every result is the product of four AI agents working concurrently, each producing a sourced report. ResumeGuard assembles their findings into a single weighted trust index.

Work-history agent: Verifies employers, titles, timelines against public records, news, and registry data.
Education agent: Confirms institutions, degrees, and graduation years against accreditation bodies.
Skills agent: Cross-checks claimed skills against role history and certifications for internal consistency.
Certifications agent: Validates issuing bodies, certificate numbers, and expiry dates.

Chapter III

Lodging a resume

ResumeGuard accepts PDF, DOCX, and plain text. Files up to 10 MB. Multi-column layouts and non-English copy are handled by embedded OCR.

NoticeReject inputs > 10 MB or corrupted OCR scans — the intake clerk will not lodge them.

Chapter IV

Reading the trust index

The trust index runs 0 – 100. Bands have plain meanings, printed on every result:

BAND	RANGE	READING
High confidence	85 – 100	Claims supported by multiple independent sources.
Standard	60 – 84	Most claims verified; minor gaps noted.
Flagged for review	40 – 59	Material inconsistencies require human review.
Unreliable	0 – 39	Significant evidence of fabrication.

Chapter V

ResumeGuard API reference

All endpoints accept and return JSON. Authenticate with a bearer token. Rate limited at 60 requests per minute per key.

POST

/api/v1/document

Lodge a single resume (multipart/form-data with file field 'resume').

POST

/api/v1/document/bulk

Lodge multiple resumes in bulk (multipart/form-data with multiple 'resumes' fields, optional 'category').

GET

/api/v1/document

List lodged documents with optional pagination (?page=1&limit=10).

GET

/api/v1/document/:id

Retrieve a document profile, parsed JSON metadata, and download link.

PUT

/api/v1/document/:id/data

Manually correct/update parsed resume JSON schema data.

DELETE

/api/v1/document/:id

Permanently delete a document and its analyses.

GET

/api/v1/result/:id

Retrieve verification result (multi-agent score, findings, checks).

POST

/api/v1/result/:id/rerun

Re-run the AI multi-agent verification scan on the document.

POST

/api/v1/verify/item/:id

Request validation check on a specific work experience or education index.

PUT

/api/v1/document/:id/override

Override a specific finding audit status ('correct' or 'fraud').

POST

/api/v1/deep-analysis

Initiate deep background search on key experiences.

Chapter VI

Billing & credits

One credit equals one verified result. Re-runs on the same file within 30 days are free. Credits do not roll over to the next month. Upgrade in-flight; unused balance carries forward pro-rata.

Chapter VII

Security & retention

TLS 1.3 in transit; AES-256 at rest. SOC 2 Type II compliant. Files are retained for the duration of your subscription and purged within 30 days of cancellation. Enterprise contracts may elect alternative residency and bring-your-own-key encryption.