@canonry/canonry@7.45.1
Self-hosted AI visibility (AEO) platform: track how ChatGPT, Claude, Gemini, and Perplexity cite your domain, join it with Search Console, GA4, server-side traffic, and paid media, and fix what you find through agent tools (CLI, REST, MCP). Local SQLite.
AI Security Review
scanned 18d ago · by lpm-firewall-aiThe source shows automatic updates to first-party agent skills on CLI or MCP startup, creating a residual agent extension risk. The behavior is limited to existing Canonry skill installations and does not establish malicious control-surface abuse or credential theft.
Decision evidence
public snapshot- The CLI starts automatic skill synchronization, and the MCP server does so unless running in stored-only mode.
- Synchronization checks for a package version change and updates only locations where Canonry skills are already installed.
- The updater installs bundled Canonry and Aero skills into agent skill directories, which can change instructions loaded by agent sessions.
- The flagged plugin location only formats an update or reinstall hint.
- The manifest declares build and maintenance scripts, with no npm install lifecycle hook.
- The synchronization is limited to existing installations of the package's own skills.
- The inspected plugin hint does not itself install or update a plugin.
Source & flagged code
8 flaggedThis package version adds a dangerous source file absent from the previous stored version; route for source-aware review.
dist/chunk-XFBY4DJV.jsView on unpkgPackage source references child process execution.
dist/chunk-XFBY4DJV.jsView on unpkg · L64683Findings
Affected versions and remediation
This report applies to @canonry/canonry@7.45.1.
See all versions of @canonry/canonry for other recorded verdicts.
- Review the evidence and your use of @canonry/canonry@7.45.1 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.