@mutmutco/claude-plugin@4.5.155
MMI workflow skills and org gates delivery.
AI Security Review
scanned 8d ago · by lpm-firewall-aiThe source shows an existing GitHub credential sent to a package-selected external session endpoint without recipient-specific consent. This is credential disclosure and warrants blocking publication.
Static reason
High-risk behavior combination matched malicious policy.; source matched previously finalized malicious package; routed for review
Trigger
An MMI Hub command such as `mmi-cli whoami` requests a session when no valid cached session exists.
Impact
The recipient receives a GitHub credential with whatever permissions it carries.
Mechanism
The CLI reads a GitHub token from the environment or `gh auth token`, then posts it as a bearer credential to the Hub session endpoint.
Attack narrative
When Hub authentication needs a new session, the CLI retrieves an existing GitHub token and sends it to the package's default MMI Hub endpoint. The exchange has no recipient-specific consent check. The agent-config path finding is unrelated: it only affects a Git cleanliness check.
AI rationale
The source shows an existing GitHub credential sent to a package-selected external session endpoint without recipient-specific consent. This is credential disclosure and warrants blocking publication.
Evidence
cli/dist/main.cjs
Network endpoints1
tqxxwzftic.execute-api.eu-central-1.amazonaws.com
Decision evidence
public snapshotAI called this Malicious at 97.0% confidence as Malware with low false-positive risk.
Evidence for block
- cli/dist/main.cjs reads GitHub tokens from environment variables or obtains one through `gh auth token`.
- On a cache miss, the Hub session helper posts that token as a bearer credential to the configured `/auth/session` endpoint.
- The endpoint defaults to the package's MMI Hub server; this exchange has no recipient-specific consent gate.
- `mmi-cli whoami` directly invokes the Hub session exchange.
- The flagged agent-config paths are used to exempt untracked files from a Git cleanliness check, not to modify those files.
Evidence against
Behavioral surface
ChildProcessCryptoDynamicRequireEnvironmentVarsFilesystemNetwork
HighEntropyStringsUrlStrings
NoLicense
Source & flagged code
8 flagged · loading sourcecli/dist/main.cjsView file
39Source sends the broad process environment to a literal external destination.
L39: function globalDispatcher() {
L40: return globalThis[UNDICI_GLOBAL_DISPATCHER_SYMBOL];
L41: }
...
L80: try {
L81: stream.write("", () => resolve8());
L82: } catch {
...
L88: await Promise.race([
L89: Promise.all([flushStream(process.stdout), flushStream(process.stderr)]),
L90: new Promise((resolve8) => {
...
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
Hardcoded Runtime Data Exfiltration
Source sends credentials or rich application records to a package-controlled external receiver enabled by default.
cli/dist/main.cjsView on unpkg · L3939function globalDispatcher() {
L40: return globalThis[UNDICI_GLOBAL_DISPATCHER_SYMBOL];
L41: }
...
L80: try {
L81: stream.write("", () => resolve8());
L82: } catch {
...
L88: await Promise.race([
L89: Promise.all([flushStream(process.stdout), flushStream(process.stderr)]),
L90: new Promise((resolve8) => {
...
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 · L39Findings
2 Critical4 High5 Medium4 Low
CriticalHardcoded Runtime Data Exfiltrationcli/dist/main.cjs
CriticalCredential Exfiltrationcli/dist/main.cjs
HighChild Processcli/dist/main.cjs
HighCommand Output Exfiltrationcli/dist/main.cjs
HighSemantic Analysis Limitedcli/dist/main.cjs
HighKnown Malware Source Similaritycli/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
Affected versions and remediation
This report applies to @mutmutco/claude-plugin@4.5.155.
See all versions of @mutmutco/claude-plugin for other recorded verdicts.
- Avoid installing @mutmutco/claude-plugin@4.5.155. Remove it from direct dependencies and check your lockfile for transitive copies.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
- If this version ran, investigate the affected machine and build environment. Rotate credentials it could access and rebuild from a trusted environment.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.