registry / @canonry/canonry / 7.64.5
@canonry/canonry@7.64.5
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 10d ago · by lpm-firewall-aiSource inspection confirms a first-party agent-skill update capability that runs automatically after prior installation, so a warning is appropriate for the residual risk of changed agent instructions. The package has no install hook, and the reviewed source did not establish credential theft, broad agent configuration changes, or another concrete attack.
Static reason
High-risk behavior combination matched malicious policy.
Trigger
A non-stored-only CLI command or MCP startup when an eligible managed skill exists and the package version changes or its sync interval elapses.
Impact
Agent sessions can receive changed Canonry-authored instructions without a fresh confirmation. The inspected behavior is scoped to these skills and shows no credential transfer or broad agent settings changes.
Mechanism
The sync copies updated bundled files into Canonry skill directories and maintains Codex symlinks when present. It preserves locally edited files unless force is requested.
AI rationale
Source inspection confirms a first-party agent-skill update capability that runs automatically after prior installation, so a warning is appropriate for the residual risk of changed agent instructions. The package has no install hook, and the reviewed source did not establish credential theft, broad agent configuration changes, or another concrete attack.
Evidence
dist/chunk-56ISY5RC.jsdist/cli.jsdist/mcp.jsdist/init-X2ADJGSC.jsdist/chunk-FBHW62IM.jsdist/chunk-T3QZVIPV.js.claude/skills/canonry.claude/skills/aero.codex/skills/canonry.codex/skills/aero
Decision evidence
public snapshotAI called this Suspicious at 84.0% confidence as Dangerous Capability with medium false-positive risk.
Evidence for warning
- package.json lists CLI and library entrypoints and build scripts, with no preinstall, install, or postinstall hook.
- The explicit init flow installs bundled Canonry and Aero skills unless skipped; the skill installer writes their managed files and manifest.
- CLI startup and non-stored-only MCP startup call the automatic sync. It runs after a package version change or the default 24-hour interval, with an environment opt-out.
- Automatic sync only selects Canonry skill directories in the home or current project when their managed manifest exists, then refreshes those skills.
- The sync compares file hashes with the prior manifest. It refreshes unchanged managed files and preserves locally edited files unless force is requested.
Evidence against
- The behavior is limited to Canonry-owned skill directories and preserves local edits; the inspected sync path does not change broad agent settings or transfer credentials.
Behavioral surface
ChildProcessCryptoEnvironmentVarsEval
Source & flagged code
8 flaggeddist/chunk-PKRH4UKJ.jsView file
59954// ../integration-commoncrawl/src/plugin-installer.ts
L59955: import { spawn } from "child_process";
L59956: import fs4 from "fs/promises";
High
Child Process
Package source references child process execution.
dist/chunk-PKRH4UKJ.jsView on unpkg · L599541190});
L1191: return `${PREFIX}${salt.toString("base64")}$${derived.toString("base64")}`;
L1192: }
...
L1497: });
L1498: app.post("/auth/login", async (request, reply) => {
L1499: const parsed = loginRequestSchema.safeParse(request.body);
...
L4958: try {
L4959: const parsed = new URL(/^https?:\/\//i.test(trimmed) ? trimmed : `https://${trimmed}`);
L4960: return parsed.hostname.toLowerCase().replace(/\.$/, "") || null;
...
L6691: if (first === 10 || first === 172 && second >= 16 && second <= 31 || first === 192 && second === 168) {
L6692: return "which is private space";
L6693: }
High
Command Output Exfiltration
Source combines command execution, command-output handling, and outbound requests; review data flow before blocking.
dist/chunk-PKRH4UKJ.jsView on unpkg · L1190•stage = ast_semantic_analysis; reason = ast_path_work_budget_exceeded; limitedFiles = 2
Findings
3 Critical5 High3 Medium5 Low
CriticalBrowser Credential Phishingdist/chunk-T3QZVIPV.js
CriticalTrojan Source Unicodeassets/assets/vendor-yaml-Y43rmNgW.js
CriticalTrigger Reachable Dangerous Capabilitydist/chunk-T3QZVIPV.js
HighChild Processdist/chunk-PKRH4UKJ.js
HighCredential Redirect Persistencedist/chunk-LEGBODMN.js
HighCommand Output Exfiltrationdist/chunk-PKRH4UKJ.js
HighCross File Remote Execution Contextdist/chunk-YXXJFOVG.js
HighSemantic Analysis Limiteddist/chunk-PKRH4UKJ.js
MediumNetwork
MediumEnvironment Vars
MediumStructural Risk Force Deep Review
LowScripts Present
Low
Affected versions and remediation
This report applies to @canonry/canonry@7.64.5.
See all versions of @canonry/canonry for other recorded verdicts.
- Review the evidence and your use of @canonry/canonry@7.64.5 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.