momentic-mobile@1.36.0
The official CLI for Momentic Mobile. Build, run, and manage end-to-end tests for iOS and Android through natural language prompts.
AI Security Review
scanned 1mo ago · by lpm-firewall-aiThe inspected source supports a warning for explicit CLI agent configuration capability, not a malicious install-time attack. Routine skills updates require confirmation, while explicit noninteractive initialization can configure agents without prompts.
Static reason
One or more suspicious static signals were detected.; previous stored version diff introduced dangerous source
Trigger
Explicit skills, initialization, or MCP setup commands; confirmed skills prompts during routine commands.
Impact
Changes future agent instructions and available tools; noninteractive initialization can apply setup without a prompt.
Mechanism
Writes package-provided skills and instruction pointers, saves target preferences, and configures agents to launch the Momentic MCP server.
AI rationale
The inspected source supports a warning for explicit CLI agent configuration capability, not a malicious install-time attack. Routine skills updates require confirmation, while explicit noninteractive initialization can configure agents without prompts.
Evidence
bin/cli.jsAGENTS.mdCLAUDE.md.momentic/skills.json
Decision evidence
public snapshotAI called this Suspicious at 90.0% confidence as Dangerous Capability with medium false-positive risk.
Evidence for warning
- package.json exposes bin/cli.js as the CLI entrypoint and declares no install lifecycle hooks.
- CLI skills setup targets agent skill directories and AGENTS.md instruction pointers.
- The skills installer writes generated instruction content into agent pointer files.
- Routine command setup prompts before installing or updating skills; saved targets do not bypass confirmation.
- MCP configuration launches the Momentic CLI and can include an explicitly selected API key.
- Explicit initialization can install skills and MCP configuration without prompts in noninteractive mode.
Evidence against
- The indicated agent location is a project-search exclusion list.
- Inspected source did not corroborate the claimed fixed unofficial credential gateway or command-output exfiltration.
- Agent configuration changes occur through CLI commands rather than npm install hooks.
Behavioral surface
ChildProcessCryptoDynamicRequireEnvironmentVarsEvalFilesystem
Source & flagged code
12 flaggedbin/cli.jsView file
•matchType = previous_version_dangerous_delta
matchedPackage = momentic-mobile@1.35.1
matchedIdentity = npm:bW9tZW50aWMtbW9iaWxl:1.35.1
similarity = 0.636
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.
bin/cli.jsView on unpkg32import { PostHog } from 'posthog-node';
L33: import * as rst from 'child_process';
L34: import { execFile, execSync, execFileSync, exec } from 'child_process';
High
181L182: What the agent did: ${r}`:e}var Ow=Ga.extend({type:z.literal("AI_ACTION_DYNAMIC"),text:z.string(),effort:vt.optional(),retries:z.number().optional(),version:Pw.optional(),cache:RF....
L183: `).map((r,n)=>n>0?`${o}${r}`:r).join(`
...
L189: `)}`;}o.required===false&&(n=n.nullable(),i+=`
L190: Use null when unknown or not applicable.`),e[o.name]=n.describe(i);}return z.strictObject(e)}function oNe(t){let e=[t.testCases??[],t.bugReproductions??[]].map(o=>z.toJSONSchema(tN...
L191: `)){let[r,n]=o.trim().split(/\s+/,2);if(!r||!n)continue;let i=Number.parseInt(n,10);Number.isFinite(i)&&(e[r]=i);}return e}function JBe(t,e){let o,r,n;for(let i of e.split(`
L192: `)){let a=i.indexOf(":");if(a===-1)continue;let s=i.slice(0,a),l=i.slice(a+1).trim();if(s==="Name"&&(o=l),s==="PPid"&&(r=Numb
Findings
1 Critical10 High4 Medium6 Low
CriticalPrevious Version Dangerous Deltabin/cli.js
HighChild Processbin/cli.js
HighShell
HighSame File Env Network Executionbin/cli.js
HighCredential Redirect Persistencebin/cli.js
HighCommand Output Exfiltrationbin/cli.js
HighSandbox Evasion Gated Capabilitybin/cli.js
HighCloud Metadata Accessbin/cli.js
HighTrigger Reachable Command Output Exfiltrationbin/cli.js
HighTrigger Reachable External Post Callbackbin/cli.js
HighSemantic Analysis Limitedbin/cli.js
MediumDynamic Requirebin/cli.js
Medium
Affected versions and remediation
This report applies to momentic-mobile@1.36.0.
See all versions of momentic-mobile for other recorded verdicts.
- Review the evidence and your use of momentic-mobile@1.36.0 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.