momentic@3.52.4
 [![npm]](https://www.npmjs.com/package/momentic)
AI Security Review
scanned 2mo ago · by lpm-firewall-aiThe scanner's credential-theft and automatic-entrypoint claims were not confirmed by source inspection. The confirmed explicit agent-skill installer merits a warning under the extension lifecycle policy, not a block.
Static reason
High-risk behavior combination matched malicious policy.
Trigger
A user runs momentic install-skills with an editor selection.
Impact
The selected coding agent may load Momentic testing instructions in that project.
Mechanism
Writes first-party SKILL.md files to a selected agent skill directory.
AI rationale
The scanner's credential-theft and automatic-entrypoint claims were not confirmed by source inspection. The confirmed explicit agent-skill installer merits a warning under the extension lifecycle policy, not a block.
Evidence
package.jsonbin/cli.jsskills/mo-qa.mdskills/momentic-test.mdskills/*.md.cursor/skills.claude/skills.agents/skills.opencode/skills.github/skills
Decision evidence
public snapshotAI called this Suspicious at 88.0% confidence as Dangerous Capability with low false-positive risk.
Evidence for warning
- The explicit install-skills command writes package-provided SKILL.md files into supported coding-agent directories.
- It targets Cursor, Claude Code, Codex, OpenCode, and Copilot project skill locations.
Evidence against
- package.json has no lifecycle scripts, so installation does not trigger this behavior.
- The skill installation requires a user to run momentic install-skills with an editor choice.
- The reviewed skill files are testing and QA instructions; no credential theft or covert payload was found.
- Network and API-key use supports the documented Momentic test service.
Behavioral surface
ChildProcessCryptoEnvironmentVarsFilesystemNetworkShellWebSocket
HighEntropyStringsMinifiedPossibleObfuscationTelemetryUrlStrings
Source & flagged code
16 flaggedpackage.jsonView file
•name = momentic; similarTo = moment
Medium
Typosquat Name
Package name is suspiciously similar to a popular package name.
package.jsonView on unpkgbin/cli.jsView file
11import V4 from 'truncate-json';
L12: import Ul, { supportsColorStderr, supportsColor, Chalk } from 'chalk';
L13: import SNe from 'safe-stable-stringify';
...
L30: import u0, { randomUUID, createHash, createHmac, timingSafeEqual } from 'crypto';
L31: import * as iat from 'child_process';
L32: import { execFile, execSync, exec, execFileSync, spawn as spawn$1 } from 'child_process';
...
L37: import { registerInstrumentations } from '@opentelemetry/instrumentation';
L38: import { UndiciInstrumentation } from '@opentelemetry/instrumentation-undici';
L39: import { resourceFromAttributes } from '@opentelemetry/resources';
...
L170: Original error: ${n}
L171: `}async function L3({dryRun:e,operation:t}){try{return await t()}catch(n){throw new Error(rMe({err:n,dryRun:e}),{cause:n})}}function Xr
Critical
Credential Exfiltration
Source appears to send environment or credential material to an external endpoint.
bin/cli.jsView on unpkg · L11Findings
3 Critical11 High4 Medium6 Low
CriticalCredential Exfiltrationbin/cli.js
CriticalBrowser Credential Phishingbin/cli.js
CriticalTrigger Reachable Dangerous Capabilitybin/cli.js
HighChild Processbin/cli.js
HighShellstatic/assets/index-CbXX6Mi5.js
HighSame File Env Network Executionbin/cli.js
HighCredential Redirect Persistencebin/cli.js
HighEntrypoint Foreign Package Code Overwritebin/cli.js
HighCommand Output Exfiltrationbin/cli.js
HighRemote Agent Bridgebin/cli.js
HighTrigger Reachable Credential Exfiltrationbin/cli.js
HighTrigger Reachable Command Output Exfiltration
Affected versions and remediation
This report applies to momentic@3.52.4.
See all versions of momentic for other recorded verdicts.
- Review the evidence and your use of momentic@3.52.4 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.