registry / autopilot-ai / 3.1.0
autopilot-ai@3.1.0
autopilot (formerly kimiflare): a terminal coding agent for any model with your own key: OpenRouter, Anthropic, OpenAI, Gemini, Bedrock, Vertex, Ollama and more. Per-turn cost, web search, GitHub, headless browser, LSP, MCP, local memory.
AI Security Review
scanned 6d ago · by lpm-firewall-aiThe flagged DNS code checks web-fetch destinations, and provider credentials are associated with their provider endpoints. The inbox sends its app-specific secret only after the user submits its form; no automatic credential exfiltration or unrelated execution was identified.
Static reason
High-risk behavior combination matched malicious policy.; previous stored version diff introduced dangerous source
Trigger
The user starts the CLI, requests an agent task, uses web fetch, or opens the inbox feature.
Impact
Agent prompts and tool results may reach the selected model provider, and the inbox sends its submitted handle and secret to its service; no unrelated malicious impact was identified.
Mechanism
The app routes provider keys to the configured model service, checks web-fetch destination addresses, and sends inbox credentials only after form submission.
AI rationale
The flagged DNS code checks web-fetch destinations, and provider credentials are associated with their provider endpoints. The inbox sends its app-specific secret only after the user submits its form; no automatic credential exfiltration or unrelated execution was identified.
Network endpoints3
api.openai.comai-gateway.vercel.shhello.kimiflare.com
Decision evidence
public snapshotAI called this Clean at 95.0% confidence as Benign with low false-positive risk.
Evidence for AI clean decision
- The manifest exposes a CLI and SDK; its prepublish hook runs the build.
- Provider presets pair OpenAI keys with api.openai.com and use a separate AI_GATEWAY_API_KEY for Vercel AI Gateway.
- The DNS lookup code resolves requested web-fetch targets and classifies their addresses.
- The inbox asks for a handle and app-specific secret, then sends them to hello.kimiflare.com after the user submits the form.
Evidence against
- The coding agent can execute local commands and send requested prompts and tool results to the selected model provider; these capabilities fit the advertised CLI purpose.
Behavioral surface
ChildProcessCryptoDynamicRequireEnvironmentVarsEvalFilesystemNetworkShell
HighEntropyStringsUrlStrings
Source & flagged code
15 flaggeddist/index.jsView file
1324const iat = Math.floor(now2 / 1e3);
L1325: const payload = enc({ iss: sa.client_email, scope: SCOPE, aud: sa.token_uri ?? "https://oauth2.googleapis.com/token", iat, exp: iat + 3600 });
L1326: const signature = createSign("RSA-SHA256").update(`${header}.${payload}`).sign(sa.private_key).toString("base64url");
...
L1331: const tokenUrl = creds.type === "service_account" ? creds.token_uri ?? "https://oauth2.googleapis.com/token" : "https://oauth2.googleapis.com/token";
L1332: const body = creds.type === "service_account" ? new URLSearchParams({ grant_type: "urn:ietf:params:oauth:grant-type:jwt-bearer", assertion: signJwt(creds, now2) }) : new URLSearchP...
L1333: const res = await fetchImpl2(tokenUrl, { method: "POST", headers: { "Content-Type": "application/x-www-form-urlencoded" },
Critical
Command Output Exfiltration
Source executes local commands and sends command output to an external endpoint.
dist/index.jsView on unpkg · L132439function home(env) {
L40: return env.HOME || homedir();
L41: }
...
L50: }
L51: function configDir(env = process.env) {
L52: return join(configHome(env), APP_DIR_NAME);
...
L320: if (currentStream) {
L321: currentStream.end();
L322: currentStream = null;
...
L452: function setFetchForTesting(f) {
L453: fetchImpl = f ?? ((url, init) => fetch(url, init));
L454: }
Critical
Dns Exfiltration
Source appears to send environment or credential material through DNS lookups.
dist/index.jsView on unpkg · L39Findings
6 Critical8 High4 Medium6 Low
CriticalHardcoded Runtime Data Exfiltrationdist/sdk/index.js
CriticalCommand Output Exfiltrationdist/index.js
CriticalDns Exfiltrationdist/index.js
CriticalRemote Response Code Executiondist/index.js
CriticalTrigger Reachable Dangerous Capabilitydist/index.js
CriticalPrevious Version Dangerous Deltadist/index.js
HighChild Processdist/index.js
HighShelldist/index.js
HighEval
HighSame File Env Network Executiondist/index.js
HighCredential Redirect Persistencedist/index.js
HighRemote Agent Bridgedist/index.js
Affected versions and remediation
This report applies to autopilot-ai@3.1.0.
See all versions of autopilot-ai for other recorded verdicts.
- Review the evidence and your use of autopilot-ai@3.1.0 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.