Claude Security scans code, suggests patches
Anthropic’s official Claude Code plugin maps repositories, models threats, investigates vulnerabilities with multiple agents, and independently verifies findings before producing targeted patch files for review. It runs inside the developer’s session under existing permissions, using the Claude models available to that account.
Claude Security is a compelling second-pass AppSec teammate, but its probabilistic workflow makes human review and defense-in-depth essential.
- –Cross-file reasoning and data-flow tracing target business-logic flaws that pattern-based SAST tools often miss.
- –Independent verifiers discard candidates they cannot confirm, though repeated scans can produce different results because the process is nondeterministic.
- –Patches are developed in a scratch copy and reviewed before output; nothing is automatically applied, committed, or pushed.
- –The plugin complements SAST, dependency scanning, and code review rather than replacing them.
- –Early hands-on testing suggests substantial token consumption, making targeted scans before releases or major changes more practical than constant full-repository scans.
DISCOVERED
1h ago
2026-08-31
PUBLISHED
2h ago
2026-08-31
RELEVANCE
AUTHOR
AICodeKing