registry / @mutmutco/claude-plugin / 4.5.105
@mutmutco/claude-plugin@4.5.105
MMI workflow skills and org gates delivery.
AI Security Review
scanned 5d ago · by lpm-firewall-aiThe manifest has no install lifecycle hook, and the reviewed agent-path code only classifies paths. The observed command telemetry is limited to product usage metadata sent to the configured Hub, so the scanner hints do not establish malicious behavior.
Static reason
High-risk behavior combination matched malicious policy.; previous stored version diff introduced dangerous source
Trigger
A user runs a CLI command; telemetry is skipped when MMI_TELEMETRY is set to off.
Impact
No credential disclosure, agent configuration mutation, or other concrete attack behavior was established.
Mechanism
The CLI posts the command path, version, CI status, and operating system to its configured Hub endpoint. The flagged agent-path helper only filters selected paths when evaluating untracked files.
AI rationale
The manifest has no install lifecycle hook, and the reviewed agent-path code only classifies paths. The observed command telemetry is limited to product usage metadata sent to the configured Hub, so the scanner hints do not establish malicious behavior.
Decision evidence
public snapshotAI called this Clean at 88.0% confidence as Benign with low false-positive risk.
Evidence for AI clean decision
- package.json declares no lifecycle scripts and packages the CLI and plugin files.
- The flagged agent-path logic recognizes configuration and session directories to exclude them from untracked-file blocking; it does not write agent configuration.
- CLI telemetry sends the command path, package version, CI status, and operating system, and can be disabled with MMI_TELEMETRY=off.
- Telemetry is sent to the configured MMI Hub endpoint, with a package default; the CLI calls it for a parsed command.
Evidence against
Behavioral surface
ChildProcessCryptoDynamicRequireEnvironmentVarsFilesystemNetwork
HighEntropyStringsUrlStrings
NoLicense
Source & flagged code
7 flagged · loading sourcecli/dist/main.cjsView file
39function globalDispatcher() {
L40: return globalThis[UNDICI_GLOBAL_DISPATCHER_SYMBOL];
L41: }
...
L80: try {
L81: stream.write("", () => resolve7());
L82: } catch {
...
L88: await Promise.race([
L89: Promise.all([flushStream(process.stdout), flushStream(process.stderr)]),
L90: new Promise((resolve7) => {
...
L140: CLI_EXIT_WATCHDOG_MS = (() => {
L141: const testOverride = Number(process.env.MMI_CLI_TEST_EXIT_WATCHDOG_MS);
L142: return process.env.NODE_ENV === "test" && Number.isFinite(testOverride) && testOverride >= 25 ? testOverride : 1e4;
Critical
Credential Exfiltration
Source appears to send environment or credential material to an external endpoint.
cli/dist/main.cjsView on unpkg · L39•matchType = previous_version_dangerous_delta
matchedPackage = @mutmutco/claude-plugin@4.5.84
matchedIdentity = npm:QG11dG11dGNvL2NsYXVkZS1wbHVnaW4:4.5.84
similarity = 1.000
summary = stored previous version shares package body but lacks this dangerous source file
Critical
Previous Version Dangerous Delta
This package version adds a dangerous source file absent from the previous stored version; route for source-aware review.
cli/dist/main.cjsView on unpkg1111if (process.platform === "win32") {
L1112: (0, import_node_child_process2.execFileSync)("taskkill", ["/PID", String(pid), "/T", "/F"], { stdio: "ignore", windowsHide: true });
L1113: } else {
Findings
2 Critical3 High5 Medium4 Low
CriticalCredential Exfiltrationcli/dist/main.cjs
CriticalPrevious Version Dangerous Deltacli/dist/main.cjs
HighChild Processcli/dist/main.cjs
HighCommand Output Exfiltrationcli/dist/main.cjs
HighSemantic Analysis Limitedcli/dist/main.cjs
MediumDynamic Requirecli/dist/index.cjs
MediumNetwork
MediumEnvironment Vars
MediumShips Build Helperbin/mmi-cli.cmd
MediumStructural Risk Force Deep Review
LowFilesystem
LowHigh Entropy Strings
LowUrl Strings
LowNo License
Affected versions and remediation
This report applies to @mutmutco/claude-plugin@4.5.105.
See all versions of @mutmutco/claude-plugin for other recorded verdicts.
- Review the evidence and your use of @mutmutco/claude-plugin@4.5.105 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.