code-semantic-search

Pass

Audited by Gen Agent Trust Hub on Apr 24, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The script scripts/main.cjs executes a local utility located at .claude/tools/cli/hybrid-search.cjs using the spawn method. This is a standard implementation for delegating tasks to a sub-process within the skill's environment.
  • [DATA_EXFILTRATION]: Analysis of the source code confirms no network requests or access to sensitive credential files are performed.
  • [PROMPT_INJECTION]: The skill processes external data (user queries and source code) which presents a surface for indirect prompt injection. * Ingestion points: user-provided query strings and the contents of source code files being searched. * Boundary markers: absent from the analyzed implementation. * Capability inventory: execution of the search CLI via subprocess spawning. * Sanitization: input strings are trimmed but no semantic filtering or safety validation is performed.
  • [SAFE]: No obfuscation, persistence mechanisms, or privilege escalation patterns were detected.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 24, 2026, 09:28 AM