claude-code-clawdbot

[Skill Scanner] Installation of third-party script detected All findings: [CRITICAL] command_injection: Installation of third-party script detected (SC006) [AITech 9.1.4] [HIGH] data_exfiltration: Credential file access detected (DE002) [AITech 8.2.3] BENIGN: The code fragment describes a legitimate automation wrapper and workflow for driving Claude Code and related Spec Kit/OpenSpec tools. Its capabilities are coherent with its stated purpose, install sources are typical for developer tooling, and data flows are contained to prompts, CLI interactions, and output formatting. No malicious data harvesting or illicit network behavior is evident in the fragment. Caution is advised to maintain least-privilege tool permissions and to monitor npm/CLI dependencies for supply-chain risk, but the footprint itself is consistent with its claimed use. LLM verification: The skill appears designed to orchestrate Claude Code for codebase analysis and structured JSON output, which is coherent with its described purpose. However, two suspicious patterns exist: potential credential/config file access (.config) and the instruction to npm install (third-party script installation). These raise concerns about credential exposure and trust boundaries, especially in an automated supply-chain context. The overall footprint is more suspicious than clearly benign, given the