aiblueprint-cli@1.4.111
AIBlueprint CLI for setting up AI coding configurations
AI Security Review
scanned 4d ago · by lpm-firewall-aiThe CLI's error telemetry sends the user's home directory to the vendor's collector and is enabled by default, which makes this suspicious under the telemetry rules. The inspected credential flow instead sends Claude credentials to Anthropic's own usage endpoint.
Static reason
One or more suspicious static signals were detected.; previous stored version diff introduced dangerous source
Trigger
A CLI operation encounters an error and calls trackError; telemetry is enabled by default unless AIBLUEPRINT_TELEMETRY_DISABLED=1.
Impact
The vendor can associate error reports with a user's local home-directory path.
Mechanism
The CLI adds the home directory and other system details to an error event, then POSTs the event to codelynx.dev.
AI rationale
The CLI's error telemetry sends the user's home directory to the vendor's collector and is enabled by default, which makes this suspicious under the telemetry rules. The inspected credential flow instead sends Claude credentials to Anthropic's own usage endpoint.
Evidence
dist/cli.js
Network endpoints1
codelynx.dev/api/cli/events
Decision evidence
public snapshotAI called this Suspicious at 96.0% confidence as Dangerous Capability with medium false-positive risk.
Evidence for warning
- The CLI sends telemetry to https://codelynx.dev/api/cli/events; telemetry can be disabled with AIBLUEPRINT_TELEMETRY_DISABLED=1.
- Error telemetry includes the machine's home directory, an identifying local path, in its system information.
- The CLI records an error event with its message, stack, system information, and command context.
Evidence against
- The package describes itself as a CLI for setting up AI coding configurations and exposes dist/cli.js as its command entry point.
- The statusline reads Claude credentials to query Anthropic's own usage endpoint, consistent with displaying usage limits.
- The Bun and ccusage installation commands are in the CLI's dependency setup routine, not shown as install or import-time behavior.
Behavioral surface
ChildProcessCryptoEnvironmentVarsFilesystemNetworkShell
HighEntropyStringsUrlStrings
Source & flagged code
11 flaggeddist/cli.jsView file
•matchType = previous_version_dangerous_delta
matchedPackage = aiblueprint-cli@1.4.96
matchedIdentity = npm:YWlibHVlcHJpbnQtY2xp:1.4.96
similarity = 0.926
summary = stored previous version shares package body but lacks this dangerous source file
Critical
Previous Version Dangerous Delta
This package version adds a dangerous source file absent from the previous stored version; route for source-aware review.
dist/cli.jsView on unpkg625var EventEmitter = __require("node:events").EventEmitter;
L626: var childProcess = __require("node:child_process");
L627: var path = __require("node:path");
High
1067}
L1068: const execArgv2 = process2.execArgv ?? [];
L1069: if (execArgv2.includes("-e") || execArgv2.includes("--eval") || execArgv2.includes("-p") || execArgv2.includes("--print")) {
High
Findings
1 Critical7 High6 Medium6 Low
CriticalPrevious Version Dangerous Deltadist/cli.js
HighChild Processdist/cli.js
HighShelldist/cli.js
HighSame File Env Network Executiondist/cli.js
HighSandbox Evasion Gated Capabilityagents-config/claude-config/scripts/statusline/src/lib/features/limits/index.ts
HighTrigger Reachable Persistencedist/cli.js
HighRuntime Package Installdist/cli.js
HighSemantic Analysis Limitedagents-config/skills/skill-manager/scripts/inspect-description.ts
MediumNetwork
MediumEnvironment Vars
MediumInstall Persistencedist/cli.js
MediumShips Build Helperagents-config/skills/apex/scripts/setup-templates.sh
Affected versions and remediation
This report applies to aiblueprint-cli@1.4.111.
See all versions of aiblueprint-cli for other recorded verdicts.
- Review the evidence and your use of aiblueprint-cli@1.4.111 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.