apiblaze@0.21.7
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 explicit MCP configuration write creates a guarded agent-control capability that merits a warning. Static inspection found no lifecycle trigger, self-dependency, covert exfiltration, or destructive behavior.
Static reason
High-risk behavior combination matched malicious policy.
Trigger
A user invokes the apichat MCP-install flow or explicitly supplies the install option.
Impact
This grants the registered MCP endpoint availability to Codex; the action is user-invoked but modifies an AI-agent control surface.
Mechanism
The CLI constructs an MCP server block with a supplied endpoint and optional headers, then atomically writes it to the Codex configuration.
AI rationale
The explicit MCP configuration write creates a guarded agent-control capability that merits a warning. Static inspection found no lifecycle trigger, self-dependency, covert exfiltration, or destructive behavior.
Evidence
dist/index.js~/.codex/config.toml
Decision evidence
public snapshotAI called this Suspicious at 88.0% confidence as Dangerous Capability with low false-positive risk.
Evidence for warning
- The package exposes a CLI entrypoint and has no install lifecycle hook in its manifest.
- An explicit install option can add an MCP server block, including optional API-key headers, to the Codex configuration.
- The installer rejects an existing same-name server that is not an APIblaze proxy and skips non-interactive installation unless explicitly requested.
Evidence against
- No automatic install-time execution or self-dependency was found.
- No confirmed covert exfiltration or destructive action was identified.
Behavioral surface
ChildProcessCryptoEnvironmentVarsFilesystemNetworkShellWebSocket
HighEntropyStringsPossibleObfuscationUrlStrings
Source & flagged code
7 flaggeddist/index.jsView file
77const raw = fs.readFileSync(CREDENTIALS_PATH, "utf-8");
L78: return JSON.parse(raw);
L79: } catch {
...
L83: function getAccessToken() {
L84: const fromEnv = process.env.APIBLAZE_TOKEN;
L85: if (fromEnv && fromEnv.trim()) return fromEnv.trim();
...
L104: path = __toESM(require("path"));
L105: APIBLAZE_DIR = path.join(os.homedir(), ".apiblaze");
L106: CREDENTIALS_PATH = path.join(APIBLAZE_DIR, "credentials.json");
...
L116: agentCall: () => agentCall,
L117: bothDoorsBody: () => bothDoorsBody,
L118: checkProxyName: () => checkProxyName,
Critical
Credential Exfiltration
Source appears to send environment or credential material to an external endpoint.
dist/index.jsView on unpkg · L7777Trigger-reachable chain: manifest.main -> dist/index.js
L77: const raw = fs.readFileSync(CREDENTIALS_PATH, "utf-8");
L78: return JSON.parse(raw);
L79: } catch {
...
L83: function getAccessToken() {
L84: const fromEnv = process.env.APIBLAZE_TOKEN;
L85: if (fromEnv && fromEnv.trim()) return fromEnv.trim();
...
L104: path = __toESM(require("path"));
L105: APIBLAZE_DIR = path.join(os.homedir(), ".apiblaze");
L106: CREDENTIALS_PATH = path.join(APIBLAZE_DIR, "credentials.json");
...
L116: agentCall: () => agentCall,
L117: bothDoorsBody: () => bothDoorsBody,
L118: checkProxyName: () => checkProxyName,
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 · L77Findings
2 Critical6 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
HighTrigger Reachable Credential Exfiltrationdist/index.js
MediumNetwork
MediumEnvironment Vars
MediumStructural Risk Force Deep Review
LowNon Install Lifecycle Scripts
LowScripts Present
LowFilesystem
Affected versions and remediation
This report applies to apiblaze@0.21.7.
See all versions of apiblaze for other recorded verdicts.
- Review the evidence and your use of apiblaze@0.21.7 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.