registry / @mushi-mushi/cli / 0.30.1
@mushi-mushi/cli@0.30.1
Mushi CLI: SDK setup, MCP wiring, bug triage. The bug mediator for AI-built apps: plain-English diagnosis + a ready fix, in your editor.
AI Security Review
scanned 22d ago · by lpm-firewall-aiThe manifest has no install lifecycle hook, and the inspected credential transfers go to the resolved Mushi API endpoint as part of setup. Test-report sending and wizard-triggered MCP wiring have user-facing guards; no concrete attack behavior was found.
Static reason
High-risk behavior combination matched malicious policy.
Trigger
Running the setup or connection commands; sending a test report requires confirmation by default.
Impact
No malicious impact was established from the inspected source.
Mechanism
The CLI sends its SDK key and project ID to the resolved Mushi API endpoint. Connection setup can write Cursor MCP configuration after user initiation or confirmation.
AI rationale
The manifest has no install lifecycle hook, and the inspected credential transfers go to the resolved Mushi API endpoint as part of setup. Test-report sending and wizard-triggered MCP wiring have user-facing guards; no concrete attack behavior was found.
Network endpoints1
dxptnwrhwsqckaftyymj.supabase.co
Decision evidence
public snapshotAI called this Clean at 95.0% confidence as Benign with low false-positive risk.
Evidence for AI clean decision
- package.json lists build, lint, test, and typecheck scripts; it declares no npm lifecycle hook.
- The indicated index.js and init.js code only checks whether MCP configuration files exist.
- The setup flow obtains an SDK key and sends it with the project ID to its resolved Mushi API endpoint.
- The test-report send is prompted by default; it proceeds without that prompt only with an explicit send option or --yes.
- Cursor MCP configuration is written during connection setup; the setup wizard asks before launching that connection, with the prompt defaulting to no.
- The inspected child-process calls retrieve the Windows user for file ACL tightening or open a URL without a shell.
Evidence against
Behavioral surface
ChildProcessCryptoEnvironmentVarsFilesystemNetworkShell
HighEntropyStringsUrlStrings
Source & flagged code
8 flaggeddist/chunk-BES6PACT.jsView file
54import { randomUUID } from "crypto";
L55: import { spawnSync } from "child_process";
L56: import { chmodSync, existsSync, mkdirSync, readFileSync, renameSync, statSync, unlinkSync, writeFileSync } from "fs";
High
Child Process
Package source references child process execution.
dist/chunk-BES6PACT.jsView on unpkg · L54dist/index.jsView file
82import * as p from "@clack/prompts";
L83: import { spawn } from "child_process";
L84: import { randomUUID } from "crypto";
...
L92: let end = endpoint.length;
L93: while (end > 0 && endpoint.charCodeAt(end - 1) === 47) end--;
L94: return endpoint.slice(0, end);
L95: }
L96: async function deviceFetch(url, init = {}) {
L97: const controller = new AbortController();
...
L109: headers: { "Content-Type": "application/json" },
L110: body: JSON.stringify(clientId ? { client_id: clientId } : {})
L111: });
Critical
Credential Exfiltration
Source appears to send environment or credential material to an external endpoint.
dist/index.jsView on unpkg · L82Findings
2 Critical6 High3 Medium4 Low
CriticalCredential Exfiltrationdist/index.js
CriticalTrigger Reachable Dangerous Capabilitydist/index.js
HighChild Processdist/chunk-BES6PACT.js
HighShelldist/index.js
HighCredential Redirect Persistencedist/index.js
HighCommand Output Exfiltrationdist/index.js
HighTrigger Reachable Credential Exfiltrationdist/index.js
HighTrigger Reachable Command Output Exfiltrationdist/index.js
MediumNetwork
MediumEnvironment Vars
MediumStructural Risk Force Deep Review
LowScripts Present
LowFilesystem
Low
Affected versions and remediation
This report applies to @mushi-mushi/cli@0.30.1.
See all versions of @mushi-mushi/cli for other recorded verdicts.
- Review the evidence and your use of @mushi-mushi/cli@0.30.1 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.