motifcode@0.2.1
Unofficial coding agent harness built for Motif-3: a Claude Code-style terminal session over the hosted endpoint
AI Security Review
scanned 3d ago · by lpm-firewall-aiThe reviewed credential flow is disclosed and tied to user-provided Infron credentials for the package’s hosted model client. I found no install hook or unrelated credential collection in the reviewed entry point and bundle.
Static reason
High-risk behavior combination matched malicious policy.
Trigger
A user runs the CLI and provides an Infron API key, or makes a model request with a configured key.
Impact
No unauthorized credential collection or unrelated data transfer was identified.
Mechanism
The client sends the provided key to its configured model endpoint for authentication. The default endpoint is llm.onerouter.pro; the endpoint can be configured.
AI rationale
The reviewed credential flow is disclosed and tied to user-provided Infron credentials for the package’s hosted model client. I found no install hook or unrelated credential collection in the reviewed entry point and bundle.
Evidence
~/.motif/.env
Network endpoints2
llm.onerouter.proinfron.ai
Decision evidence
public snapshotAI called this Clean at 91.0% confidence as Benign with low false-positive risk.
Evidence for AI clean decision
- The package ships a CLI entry point and declares no runtime dependencies.
- The README tells users to obtain an Infron API key and paste it into the CLI; it says the key is checked against the endpoint and saved locally.
- The login flow prompts for an Infron API key and explains that it will be checked against the configured endpoint before saving.
- Model requests send the supplied API key as a bearer token to the configured endpoint.
Evidence against
Behavioral surface
ChildProcessCryptoEnvironmentVarsFilesystemNetworkShell
HighEntropyStringsUrlStrings
Source & flagged code
6 flaggeddist/motif.jsView file
214} else if (ch < " ") {
L215: out += "\\u" + ch.charCodeAt(0).toString(16).padStart(4, "0");
L216: } else {
...
L1791: import { dirname, join } from "node:path";
L1792: var DEFAULT_ENDPOINT = "https://llm.onerouter.pro";
L1793: var DEFAULT_MODEL = "motif/motif-3";
...
L1820: }
L1821: function defaultDotenvPaths(cwd = process.cwd()) {
L1822: return [join(cwd, ".env"), join(homedir(), ".motif", ".env")];
...
L1828: function resolveEndpointConfig(opts = {}) {
L1829: const env = opts.env ?? process.env;
L1830: const files = [];
Critical
Credential Exfiltration
Source appears to send environment or credential material to an external endpoint.
dist/motif.jsView on unpkg · L214214Trigger-reachable chain: manifest.bin -> dist/motif.js
L214: } else if (ch < " ") {
L215: out += "\\u" + ch.charCodeAt(0).toString(16).padStart(4, "0");
L216: } else {
...
L1791: import { dirname, join } from "node:path";
L1792: var DEFAULT_ENDPOINT = "https://llm.onerouter.pro";
L1793: var DEFAULT_MODEL = "motif/motif-3";
...
L1820: }
L1821: function defaultDotenvPaths(cwd = process.cwd()) {
L1822: return [join(cwd, ".env"), join(homedir(), ".motif", ".env")];
...
L1828: function resolveEndpointConfig(opts = {}) {
L1829: const env = opts.env ?? process.env;
L1830: const files = [];
Critical
Trigger Reachable Dangerous Capability
A package entrypoint or install-time lifecycle script reaches a source file with blocking dangerous behavior.
dist/motif.jsView on unpkg · L214Findings
2 Critical1 High3 Medium6 Low
CriticalCredential Exfiltrationdist/motif.js
CriticalTrigger Reachable Dangerous Capabilitydist/motif.js
HighTrigger Reachable Credential Exfiltrationdist/motif.js
MediumNetwork
MediumEnvironment Vars
MediumStructural Risk Force Deep Review
LowFilesystem
LowHigh Entropy Strings
LowUrl Strings
LowSuspicious Dependency Evidencepackage.json
LowAi Review EvidenceREADME.md
LowAi Review Evidencedist/motif.js
Affected versions and remediation
This report applies to motifcode@0.2.1.
See all versions of motifcode for other recorded verdicts.
- Review the evidence and your use of motifcode@0.2.1 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.