registry / @canonry/canonry / 7.46.0
@canonry/canonry@7.46.0
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 18d ago · by lpm-firewall-aiSource inspection found guarded, first-party agent skill synchronization that can update existing Canonry skills during runtime. That supports a warning for agent-extension lifecycle risk, but the inspected evidence does not establish credential theft or another malicious attack.
Static reason
High-risk behavior combination matched malicious policy.; previous stored version diff introduced dangerous source
Trigger
Invoking the CLI outside stored-only mode or starting the MCP server outside stored-only scope.
Impact
Previously installed Canonry agent skills may be refreshed automatically, changing instructions available to the agent.
Mechanism
The package periodically checks for existing Canonry skill manifests and may reconcile bundled skill files in the user home or current directory; an environment setting can disable CLI synchronization.
AI rationale
Source inspection found guarded, first-party agent skill synchronization that can update existing Canonry skills during runtime. That supports a warning for agent-extension lifecycle risk, but the inspected evidence does not establish credential theft or another malicious attack.
Evidence
package.jsondist/cli.jsdist/mcp.jsdist/chunk-GWRIZSIL.jsdist/chunk-WMG365JI.js.claude/skills/.codex/skills/
Decision evidence
public snapshotAI called this Suspicious at 78.0% confidence as Dangerous Capability with medium false-positive risk.
Evidence for warning
- The package exposes CLI and MCP entrypoints, with no install lifecycle hook shown in its manifest.
- CLI and MCP startup can automatically synchronize bundled skills after invocation.
- Skill synchronization targets the user home or current directory when Canonry skill manifests already exist, and can update those skills.
- The installer writes bundled skill files under the agent skills directory; this is a first-party agent extension capability with a concrete automatic update path.
- The Typesafe integration sends its explicitly supplied API key to its named service endpoint and rejects redirects.
Evidence against
- Automatic skill synchronization is limited to detected Canonry-managed skills and has an environment opt-out; other sync paths have interval and configuration checks.
- The indicated code paths do not establish credential theft, unrelated agent configuration changes, or malicious install-time behavior.
Behavioral surface
ChildProcessCryptoEnvironmentVarsEval
Source & flagged code
8 flaggeddist/chunk-2UQ6WVWP.jsView file
7597// src/cloudflare-worker-deploy.ts
L7598: import { spawn } from "child_process";
L7599: import { randomUUID } from "crypto";
High
Child Process
Package source references child process execution.
dist/chunk-2UQ6WVWP.jsView on unpkg · L7597169Cross-file remote execution chain: dist/chunk-2UQ6WVWP.js spawns dist/chunk-V4TITINC.js; helper contains network access plus dynamic code execution.
L169: URL_PATH_NORMALIZATION_VERSION,
L170: USER_PASSWORD_MIN_LENGTH,
L171: UserRoles,
...
L562: if (!/^\d+$/.test(rawPort) || !Number.isSafeInteger(port) || port < 1 || port > 65535) {
L563: throw new CliError({ code: "INVALID_PORT", message: "Demo port must be an integer between 1 and 65535.", exitCode: 1 });
L564: }
...
L610: }
L611: const url = `http://${host.includes(":") ? `[${host}]` : host}:${port}`;
L612: const result = { status: "ready", mode: "view-only", sampleData: true, url };
L613: if (isMachineFormat(options.format ?? "text")) console.log(JSON.stringify(result));
...
L1055: try {
L1056: diff = ev.diff ? JSON.parse(ev.diff) : {}
High
Cross File Remote Execution Context
Source spawns a local helper that also contains network and dynamic execution context; review data flow before blocking.
dist/chunk-2UQ6WVWP.jsView on unpkg · L169Findings
3 Critical5 High3 Medium6 Low
CriticalBrowser Credential Phishingdist/chunk-V4TITINC.js
CriticalTrigger Reachable Dangerous Capabilitydist/chunk-V4TITINC.js
CriticalPrevious Version Dangerous Deltadist/chunk-ZOTA3MOL.js
HighChild Processdist/chunk-2UQ6WVWP.js
HighCredential Redirect Persistencedist/chunk-AYMS4LCP.js
HighCommand Output Exfiltrationdist/chunk-ZOTA3MOL.js
HighCross File Remote Execution Contextdist/chunk-2UQ6WVWP.js
HighSemantic Analysis Limiteddist/chunk-ZOTA3MOL.js
MediumNetwork
MediumEnvironment Vars
MediumStructural Risk Force Deep Review
LowScripts Present
Low
Affected versions and remediation
This report applies to @canonry/canonry@7.46.0.
See all versions of @canonry/canonry for other recorded verdicts.
- Review the evidence and your use of @canonry/canonry@7.46.0 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.