apiblaze@0.21.27
APIblaze CLI — Chat with your APIs, Manage your API keys, users and groups with the APIblaze serverless proxy
AI Security Review
scanned 1mo ago · by lpm-firewall-aiThe scanner's blocking conclusion is unsupported by the inspected activation and credential flows. Explicit agent configuration mutation and subsequent updates justify a warning under the supplied policy.
Static reason
High-risk behavior combination matched malicious policy.
Trigger
Running agent connection commands with an explicit option or accepting an interactive connection prompt.
Impact
Later agent sessions can access the configured API tools; saved consent permits subsequent configuration updates.
Mechanism
Writes Codex MCP configuration or invokes Claude MCP setup, adds project instructions, and refreshes previously authorized connections.
AI rationale
The scanner's blocking conclusion is unsupported by the inspected activation and credential flows. Explicit agent configuration mutation and subsequent updates justify a warning under the supplied policy.
Evidence
dist/index.jsconfig.tomlCLAUDE.mdAGENTS.md
Decision evidence
public snapshotAI called this Suspicious at 93.0% confidence as Dangerous Capability with medium false-positive risk.
Evidence for warning
- The manifest has no automatic installation hooks or runtime self-dependency.
- Explicit CLI setup writes MCP configuration into Codex config.toml.
- Agent setup adds persistent instructions to project CLAUDE.md or AGENTS.md files.
- Interactive setup asks which agent to connect; an explicit install option also activates setup.
- Saved installation consent enables automatic MCP configuration refresh during later CLI use.
- Agent authentication can use a newly minted tenant credential; chat optionally forwards a caller-supplied LLM key to the proxy runtime.
Evidence against
- No covert credential harvesting or malicious installation chain was established.
- Codex setup refuses to replace a conflicting server belonging to another provider.
- The inspected server client sends caller-supplied credentials to configured API service endpoints.
Behavioral surface
ChildProcessCryptoEnvironmentVarsFilesystemNetworkShellWebSocket
Source & flagged code
8 flaggeddist/index.jsView file
32L33: // package.json
L34: var version;
...
L70: function cacheFile() {
L71: return import_path.default.join(import_os.default.homedir(), ".apiblaze", "cli-policy.json");
L72: }
...
L81: for (const name of CONTROL_PLANE_ENV) {
L82: const v = process.env[name];
L83: try {
...
L120: const base2 = globalThis.fetch.bind(globalThis);
L121: globalThis.fetch = (async (input, init) => {
L122: const url = typeof input === "string" ? input : input instanceof URL ? input.href : input.url;
Critical
Credential Exfiltration
Source appears to send environment or credential material to an external endpoint.
dist/index.jsView on unpkg · L3232Trigger-reachable chain: manifest.main -> dist/index.js
L32:
L33: // package.json
L34: var version;
...
L70: function cacheFile() {
L71: return import_path.default.join(import_os.default.homedir(), ".apiblaze", "cli-policy.json");
L72: }
...
L81: for (const name of CONTROL_PLANE_ENV) {
L82: const v = process.env[name];
L83: try {
...
L120: const base2 = globalThis.fetch.bind(globalThis);
L121: globalThis.fetch = (async (input, init) => {
L122: const url = typeof input === "string" ? input : input instanceof URL ? input.href : input.url;
Critical
Trigger Reachable Dangerous Capability
A package entrypoint or install-time lifecycle script reaches a source file with blocking dangerous behavior.
dist/index.jsView on unpkg · L32Findings
2 Critical7 High3 Medium5 Low
CriticalCredential Exfiltrationdist/index.js
CriticalTrigger Reachable Dangerous Capabilitydist/index.js
HighChild Processdist/index.js
HighShell
HighSame File Env Network Executiondist/index.js
HighCredential Redirect Persistencedist/index.js
HighSandbox Evasion Gated Capabilitydist/index.js
HighCross File Remote Execution Contextdist/index.js
HighTrigger Reachable Credential Exfiltrationdist/index.js
MediumNetwork
MediumEnvironment Vars
MediumStructural Risk Force Deep Review
LowNon Install Lifecycle Scripts
Affected versions and remediation
This report applies to apiblaze@0.21.27.
See all versions of apiblaze for other recorded verdicts.
- Review the evidence and your use of apiblaze@0.21.27 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.