Evidence integrity explained
Evidence · 7 min read · Updated 8/6/2026
Remote evidence is only useful if a third party can trust it. This page explains every control applied between the moment a photo is captured and the moment a recipient verifies it.
Capture-time controls
Evidence quality is decided at capture, not afterwards. Photos taken through the guided capture flow record their own provenance as they are written.
- Device, timestamp, and — where permitted — GPS coordinates are recorded with the frame.
- Optional visible watermarking stamps time, coordinates, and workspace identity onto the image.
- Live-call captures are distinguished from self-serve uploads so reviewers know the capture method.
Verification signals
Every photo is scored against your workspace policy. Anything that fails — a screenshot, a missing timestamp, a location far from the job address — is flagged for review rather than silently accepted.
- Blur, exposure, and duplicate detection run automatically.
- Flagged photos can trigger a targeted retake request sent straight to the customer.
- Retakes keep a lineage link back to the original, so nothing disappears from the record.
Sealing
Sealing an inspection freezes it. Every photo hash, note, verdict, and metadata field is canonicalized into a single manifest, which is then signed with HMAC-SHA256.
Adding, removing, or altering anything after the seal changes the manifest, and verification will say so.
Independent verification
A sealed inspection produces a certificate link. Anyone holding it can re-run verification against the live record and download a one-page Certificate of Integrity plus machine-readable JSON.
Precise coordinates and internal notes stay hidden unless your workspace chooses to disclose them.
Frequently asked
- What happens if a photo is edited after sealing?
- Verification fails and names the specific item that no longer matches its sealed hash.
- Can a recipient verify without an account?
- Yes. Certificate links are public and require no sign-in.