registry / @canonry/canonry / 7.8.0
@canonry/canonry@7.8.0
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 28d ago · by lpm-firewall-aiThe source supports a warning for automatic updates to first-party agent skills, not the scanner’s stronger exfiltration and phishing claims. The package has no install lifecycle hook in its manifest, and the inspected agent paths show no concrete malicious behavior.
Static reason
High-risk behavior combination matched malicious policy.; previous stored version diff introduced dangerous source
Trigger
Running a CLI command that reaches the command handler, unless auto-sync is disabled or its version and interval checks skip it.
Impact
An existing Canonry skill may be updated automatically, affecting instructions available to an agent; the inspected paths do not show credential theft or a foreign control-surface write.
Mechanism
The sync checks for existing Canonry skill manifests in the home or current directory and installs or updates those skills under `.claude/skills`.
AI rationale
The source supports a warning for automatic updates to first-party agent skills, not the scanner’s stronger exfiltration and phishing claims. The package has no install lifecycle hook in its manifest, and the inspected agent paths show no concrete malicious behavior.
Evidence
dist/chunk-AAMKTL6B.jsdist/chunk-LO56QXPO.jsdist/cli.js.claude/skills
Decision evidence
public snapshotAI called this Suspicious at 82.0% confidence as Dangerous Capability with medium false-positive risk.
Evidence for warning
- The CLI automatically checks and synchronizes existing Canonry skills when its command handler runs; sync can be disabled by an environment setting.
- Synchronization is limited to installed skill targets discovered in the home or current directory, and the skill installer writes under `.claude/skills`.
- The manifest exposes CLI commands and lists build, typecheck, test, and lint scripts, with no npm install lifecycle hook.
- The inspected plugin hint recommends updating the Canonry plugin; the inspected skill check only checks for expected installed skills.
- The CLI identifies local MCP configuration paths for Claude Code, Cursor, and Codex.
Evidence against
- The inspected source shows no credential forwarding or malicious network operation in the reviewed agent-integration locations.
Behavioral surface
ChildProcessCryptoEnvironmentVarsEvalFilesystemNetwork
HighEntropyStrings
Source & flagged code
9 flaggeddist/chunk-KRFTA4I7.jsView file
885gscQueryDailyTotals: () => gscQueryDailyTotals,
L886: gscSearchData: () => gscSearchData,
L887: gscUrlInspections: () => gscUrlInspections,
...
L3872: try {
L3873: return JSON.parse(value);
L3874: } catch {
...
L4406: // ga_connections
L4407: // WARNING: private_key is authentication material; consider storing in config.yaml per AGENTS.md
L4408: `CREATE TABLE IF NOT EXISTS ga_connections (
...
L11570: try {
L11571: const parsed = new URL(/^https?:\/\//i.test(trimmed) ? trimmed : `https://${trimmed}`);
L11572: return parsed.hostname.toLowerCase().replace(/\.$/, "") || null;
Critical
Hardcoded Runtime Data Exfiltration
Source sends credentials or rich application records to a package-controlled external receiver enabled by default.
dist/chunk-KRFTA4I7.jsView on unpkg · L88560119// ../integration-commoncrawl/src/plugin-installer.ts
L60120: import { spawn } from "child_process";
L60121: import fs4 from "fs/promises";
High
Child Process
Package source references child process execution.
dist/chunk-KRFTA4I7.jsView on unpkg · L60119•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-KRFTA4I7.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-KRFTA4I7.js
CriticalBrowser Credential Phishingdist/chunk-DO7XPDG5.js
CriticalTrigger Reachable Dangerous Capabilitydist/chunk-DO7XPDG5.js
CriticalPrevious Version Dangerous Deltadist/cli.js
HighChild Processdist/chunk-KRFTA4I7.js
HighCredential Redirect Persistencedist/chunk-KRFTA4I7.js
HighCommand Output Exfiltrationdist/chunk-KRFTA4I7.js
HighCross File Remote Execution Contextdist/cli.js
HighSemantic Analysis Limiteddist/chunk-KRFTA4I7.js
MediumNetwork
MediumEnvironment Vars
MediumStructural Risk Force Deep Review
Affected versions and remediation
This report applies to @canonry/canonry@7.8.0.
See all versions of @canonry/canonry for other recorded verdicts.
- Review the evidence and your use of @canonry/canonry@7.8.0 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.