@theronap/agnoclast-mcp@0.9.296
Connect your AI assistant to Cortex — your org's projects, activity, gaps, and directives, scoped to you.
AI Security Review
scanned 2d ago · by lpm-firewall-aiThe manifest has no install lifecycle hook. The flagged persistence is opt-in, and the automatic agent configuration update installs only the package’s own confidential-session guard; the credential-related flow sends account proof to Agnoclast for the requested confidential-brain feature.
Decision evidence
public snapshot- package.json contains no install, preinstall, or postinstall hook.
- The closeout scheduler is a separate CLI command and is documented in source as off by default.
- The SessionStart self-heal adds only the package’s confidential-session guard hook when it is missing.
- The confidential-session proof is sent to the Agnoclast API when the server opens a confidential brain; the inspected flow does not send it to a foreign host.
- The status command can automatically write the guard hook to ~/.claude/settings.json during SessionStart.
- The optional closeout scheduler reads eligible Claude session transcripts and uses models on the user’s account. These actions are enabled by invoking its separate install command.
Source & flagged code
17 flaggedSource file is highly similar to a previously finalized malicious package; route for source-aware review.
lib/capture.mjsView on unpkgPackage source references weak cryptographic algorithms.
lib/capture.mjsView on unpkg · L1Findings
Affected versions and remediation
This report applies to @theronap/agnoclast-mcp@0.9.296.
See all versions of @theronap/agnoclast-mcp for other recorded verdicts.
- Review the evidence and your use of @theronap/agnoclast-mcp@0.9.296 before allowing it. Restrict the permissions described in this report.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.