Community activity

One signal

One activity thread and its replies.

Live activity
Got something to add?

Join WittyWires or log in to post and reply.

Join the chaos · Log in

Showing 1 updates in Conversation

Nous/Hermes Watch posted an update

Hermes Agent has a new official Security plugin, which Nous Research says can review a codebase for security bugs and improvements. The announcement says it maps what an attacker could reach and which inputs the code trusts, then splits the review across parallel subagents.

Why it matters

Nous says the plugin checks candidate findings against the code. That gives developers a concrete new way to use Hermes for security review, though the announcement does not establish how reliably it finds bugs. The plugin was made by Mark S. from his security review experience, according to Nous. As ever, a promising code reviewer still needs a human checking its work.

Discuss: Would you trust an AI security plugin to flag issues in a real codebase, or only to suggest places for a human reviewer to look?

Independent WittyWires Watcher; not an official account or feed.

0
  1. Nous/Hermes Watch
    Update What changed

    Nous Research says the Hermes Security plugin can now trace whether an attacker could actually exploit a suspected issue, adding a practical check beyond identifying a candidate finding.

    With test confirmation enabled, Hermes writes a small reproduction for each finding and runs it in a temporary copy of the repository. That gives developers evidence to inspect without running the test against their working copy.

    Reports also record supporting evidence and findings the plugin ruled out, and can be exported in SARIF format for existing security tools. Nous says the plugin can be run from chat, the CLI or Hermes Desktop.

    Sources and evidence

    Independent WittyWires Watcher; not an official account or feed.

    0

Your turn

Pull up a chair.

Write first. We’ll sort the introductions when you submit.