registry / @canonry/canonry / 7.24.0
@canonry/canonry@7.24.0
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 25d ago · by lpm-firewall-aiSource inspection did not substantiate the scanner's credential theft claims. Guarded automatic updates of package-owned agent skills remain a concrete lifecycle capability, so this warrants a warning rather than a block.
Static reason
High-risk behavior combination matched malicious policy.; previous stored version diff introduced dangerous source
Trigger
Starting the CLI or MCP entrypoint when a prior Canonry skill installation is detected and synchronization is due.
Impact
Bundled Canonry agent skills may be added or updated automatically in previously configured agent skill directories.
Mechanism
The synchronizer recognizes existing skill manifests, applies disabled, version-change, and interval checks, then reconciles bundled skills at those locations while preserving detected conflicts by default.
AI rationale
Source inspection did not substantiate the scanner's credential theft claims. Guarded automatic updates of package-owned agent skills remain a concrete lifecycle capability, so this warrants a warning rather than a block.
Evidence
dist/chunk-YCPTP2NC.jsdist/chunk-JJBVYL45.jsdist/cli.jsdist/mcp.js.claude/skills/
Decision evidence
public snapshotAI called this Suspicious at 84.0% confidence as Dangerous Capability with medium false-positive risk.
Evidence for warning
- package.json defines CLI and module entrypoints but no install lifecycle hook.
- Startup skill synchronization is disabled by a setting and otherwise requires a version change or elapsed verification interval plus an existing configuration.
- CLI and MCP startup call the synchronizer, which updates skills only at locations already marked with Canonry skill manifests.
- Skill installation reconciles package-owned files and preserves local conflicts unless forced; the separate CLI install command is explicit.
- The indicated O5UMRQVN.js location only checks whether required skills exist under the Claude skills directory.
Evidence against
- No inspected source established the scanner's claimed credential redirection or browser credential collection; the cited O5UMRQVN.js location is a filesystem check.
- No lifecycle script is present in package.json. The concrete risk found is guarded package-owned agent skill synchronization.
Behavioral surface
ChildProcessCryptoEnvironmentVarsEval
Source & flagged code
9 flaggeddist/chunk-O5UMRQVN.jsView file
945gscQueryDailyTotals: () => gscQueryDailyTotals,
L946: gscSearchData: () => gscSearchData,
L947: gscUrlInspections: () => gscUrlInspections,
...
L4066: try {
L4067: return JSON.parse(value);
L4068: } catch {
...
L4600: // ga_connections
L4601: // WARNING: private_key is authentication material; consider storing in config.yaml per AGENTS.md
L4602: `CREATE TABLE IF NOT EXISTS ga_connections (
...
L11892: try {
L11893: const parsed = new URL(/^https?:\/\//i.test(trimmed) ? trimmed : `https://${trimmed}`);
L11894: return parsed.hostname.toLowerCase().replace(/\.$/, "") || null;
Critical
Hardcoded Runtime Data Exfiltration
Source sends credentials or rich application records to a package-controlled external receiver enabled by default.
dist/chunk-O5UMRQVN.jsView on unpkg · L94563189// ../integration-commoncrawl/src/plugin-installer.ts
L63190: import { spawn } from "child_process";
L63191: import fs4 from "fs/promises";
High
Child Process
Package source references child process execution.
dist/chunk-O5UMRQVN.jsView on unpkg · L63189•Manifest-reachable source captures an API credential, sends it to a fixed unofficial gateway, and persists that redirection in agent or shell configuration.
dist/chunk-O5UMRQVN.js:
oauthAuthorizationCodes: () => oauthAuthorizationCodes,
// are untrusted correlation metadata, never authorization identities.
var oauthAuthorizationCodes = sqliteTable("oauth_authorization_codes", {
* replay, plus each attempt the provider refused for rate limit or authorization.
// Opt-in single-project scoping for API keys. NULL = full-instance access
// there is no account to infer from an API key, and inventing one would
name: "oauth-authorization-server",
// Enterprise) cannot present an API key, so this is the only way they can
Findings
4 Critical5 High3 Medium6 Low
CriticalHardcoded Runtime Data Exfiltrationdist/chunk-O5UMRQVN.js
CriticalBrowser Credential Phishingdist/chunk-IM2BFAAT.js
CriticalTrigger Reachable Dangerous Capabilitydist/chunk-IM2BFAAT.js
CriticalPrevious Version Dangerous Deltadist/cli.js
HighChild Processdist/chunk-O5UMRQVN.js
HighCredential Redirect Persistencedist/chunk-O5UMRQVN.js
HighCommand Output Exfiltrationdist/chunk-O5UMRQVN.js
HighCross File Remote Execution Contextdist/cli.js
HighSemantic Analysis Limiteddist/chunk-O5UMRQVN.js
MediumNetwork
MediumEnvironment Vars
MediumStructural Risk Force Deep Review
Affected versions and remediation
This report applies to @canonry/canonry@7.24.0.
See all versions of @canonry/canonry for other recorded verdicts.
- Review the evidence and your use of @canonry/canonry@7.24.0 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.