The Claude Security plugin for Claude Code is now available in beta. Scan your changes for vulnerabilities before you commit, or run a full scan across your codebase, all from your terminal on the Claude inference you already run.
Claude Code adds command-line vulnerability scanning in beta
It scans entire codebases without requiring external infrastructure.
76913.6K1.1K9.1K1.5M
Original post
Developer Impact
Checks run locally with no added infrastructure
The plugin installs through one command and then works in three layers: fast pattern matching on each edit, background diff review at the end of a turn, and deeper agentic analysis on commits, all using your existing Claude session.
Open Question
Human oversight stays mandatory and full legacy scans stay out of scope
Findings still require developer approval before any fix is applied, and the tool focuses only on Claude-generated changes rather than performing independent scans of an entire existing codebase.
Sentiment
Positive users praise Claude's new security plugin for scanning terminal code vulnerabilities before commits, while negative users object to paywalls, enterprise locks, and guardrails blocking access or tools like Fable.
Pos
72.8%
Neg
27.2%
93 comments with sentiment.
Cluster Engagement
Digg Deeper
No Digg Deeper questions have been answered for this story yet.
Posts from X
Most Activity
Most Activity
BOOKMARKS7
@claudeai i also made a security scanner skill powered by gpt-5.6 sol, so you can run security scans across different models. 100% open source :D https://github.com/agent-team-foundation/launch-readiness-scan
RETWEETS1.1K
The Claude Security plugin for Claude Code is now available in beta. Scan your changes for vulnerabilities before you commit, or run a full scan across your codebase, all from your terminal on the Claude inference you already run.
@claudeai cool, but does it get rid of this bs?
@claudeai Ironic to release this knowing your users will never be able to use it with Fable
@claudeai Will Fable fallback to Opus performing this scan?
Anthropic just brought the “Claude Security” version, ✅ Claude Code ✅ Claude Cowork ✅ Claude Design ✅ Claude Finance ✅ Claude Science ✅ Claude Teacher ✅ Human resource ✅ Development ✅ Claude Security ⬜ Claude HR ⬜ Claude Analytics ⬜ Claude Marketing ⬜ Claude Sales ⬜ Claude Legal ⬜ Claude R&D ⬜ Claude Procurement ⬜ Claude Accounting
@claudeai Love how Anthropic's first expectation of it's users is that we are malicious and we cannot be trusted, and ofc anyone wanting to learn about ML or AI in general, you poison their prompt, Evil Corp vibes.
@claudeai Nice, finally these tools are becoming more accessible 🔥
@claudeai Claude is accelerating faster
One feature I'd love from Anthropic: Make Claude subscription tokens usable through the API. If I'm paying for Pro/Max, let me spend an equivalent amount of tokens on experimentation & personal projects. Many devs want to build with Claude, not just chat with it. A shared usage pool lowers the barrier & grows the ecosystem.
@Deepusleepy @claudeai Look deep in your heart and you know the answer
@claudeai Does it work with Fable or it switch back to opus like always ?
@claudeai Why to use this when we can’t even use Fable 5 to check vulnerabilities of codebase.
@claudeai this is great, but would be even better if it didn't constantly fall back to Opus 4.8.
@claudeai FUCK CLAUDE WITH THESE MORONIC RESTRICTIONS!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! THIS APP ISN'T WORTH MY MONEY NOR MY TIME ANYMORE. THE CLAUDE DECISION MAKERS CAN GO FUCK THEMSELVES!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
─ claude plugin marketplace add anthropics/claude-plugins-official ─╯ ✔ Marketplace 'claude-plugins-official' already on disk — declared in user settings ❌ Failed to install plugin "claude-security": Plugin "claude-security" not found in any configured marketplace
@claudeai whats the point of this if i cant use it with fable. this is ur own official plugin
Load more posts
@claudeai Will it fallback to Opus for security scanning