Claim Raven glossary

Look up Claim Raven tool names and technical terms. VA claim terms are in the VA glossary at /glossary.

Raven Eye

Claim Raven's document analyzer. Upload VA letters, DBQs, or C&P exam reports for plain-English explanations with legal citations.

Ask Raven

Claim Raven's AI assistant. Answers educational questions about VA claims and shows retrieved BVA decisions, 38 CFR, M21 guidance, and record context in a source receipt when available.

Ask Raven answer depth

Ask Raven automatically adjusts how much detail an answer needs based on question complexity, record context, authorities, uncertainties, and next steps.

Raven Scan

Tool that analyzes VA Blue Button reports (and optional STRs) to surface possible claims, evidence gaps, and exposure patterns.

Raven Scribe

The account-backed drafting tool for a veteran's own personal and lay statements. Veterans can start without an account at /tools/personal-statements, then continue in Raven Scribe at /raven-scribe.

Raven Insights

VA claims intelligence dashboard with national, regional, and demographic statistics drawn from public VA data sources.

Raven Research

Legal research tool with chat and browse access to BVA decisions, 38 CFR, M21-1, and M21-5.

OAuth 2.0

A widely-used authentication protocol that lets Claim Raven verify your VA.gov identity without ever seeing your password. ID.me and Login.gov are the identity providers.

PKCE (Proof Key for Code Exchange)

An OAuth security extension that prevents authorization-code interception. Claim Raven uses PKCE on every VA.gov sign-in.

VA Lighthouse

The VA's developer API platform that powers third-party access to claims data, facilities, and benefits intake. Claim Raven is built on Lighthouse.

ICS

iCalendar file format. Plus subscribers can export their deadlines as an .ics file and import into Google Calendar, Outlook, or Apple Calendar.

PII (Personally Identifiable Information)

Information that can identify a person, such as a name, SSN, address, or VA file number. Claim Raven uses best-effort masking on supported text paths, but raw visual inputs may reach an AI provider before masking is possible.