FAQ
Short answers to the questions we hear most.
Why is an opportunity's fit score 0?
Scores are computed against your company profile. If the profile is empty, nothing can match — run the guided first run first, then refine the profile as needed.
Where do "predicted" opportunities come from?
From federal budget justification documents. WinLayer extracts the programs an agency plans to fund and scores them like any opportunity — months before a solicitation reaches SAM.gov. They're marked Predicted everywhere they appear.
Can the AI make things up?
Agents may only use the numbered evidence pack they're given, must cite it inline, and any citation to non-existent evidence is flagged as ungrounded. Thin evidence lowers confidence rather than being padded over, and risky outputs carry a review flag.
How fresh is the data?
SAM.gov syncs hourly; awards and grants nightly; protests daily; agency forecasts weekly; budget documents monthly. See Data sources.
Is there a sandbox?
Yes. Create an expiring wl_test_… API key on /developer. Sandbox requests return deterministic synthetic workspace, opportunity, pipeline, and agent data; write scopes and production identifiers are unavailable.
How do I rotate an API key?
Create a new API key on /developer, switch your integration, then revoke the old one. Revocation is immediate; the audit trail keeps the key's history.