Anthropic's terminal-based AI coding agent. Send a session log and Videlic runs all three engines — Pulse, Graph, Probe — plus a full session timeline.
What Videlic catches
Claude says it "added tests and refactored auth" — but the test file was deleted and auth.ts has a missing guard. We catch the gap.
A one-line change in a shared utility with 14 importers. Claude won't tell you about the blast radius — we will.
expect(true).toBe(true) passes every CI run but tests nothing. Videlic flags assertions that can never fail.
How to connect
Connect
Install the Videlic GitHub App and add the Claude Code hook — one line, your key included. Two minutes, once.
Open a PR
Code with Claude Code as usual. Open a PR — that's the signal to review.
Get the verdict
A clear verdict on the PR in about a minute — where Claude's claims diverge from the code, how far the change reaches, and where tests fall short.
Send your first session and get the full report — Pulse, Graph, Probe, and the timeline. Free plan, no credit card.