The Agent Skills Directory

[SAFE]: The skill instructions and metadata contain no malicious patterns, direct prompt injections, or obfuscated content.\n- [COMMAND_EXECUTION]: The skill defines workflows for executing local Python scripts (e.g., coverage_analyzer.py) and standard developer tools (npm, docker, kubectl). These operations are consistent with the skill's purpose as a QA toolkit.\n- [EXTERNAL_DOWNLOADS]: Environment setup instructions specify the use of standard, well-known package managers (npm and pip) to install dependencies.\n- [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it is designed to analyze code at a user-specified target-path (documented in SKILL.md). Evidence: 1. Ingestion point: file analysis via target-path. 2. Boundary markers: Absent. 3. Capability inventory: execution of local scripts and CLI tools. 4. Sanitization: Not specified. This risk is inherent to code analysis tools and is considered safe in this context.\n- [NO_CODE]: The core implementation logic in the scripts/ directory was not provided for analysis.

senior-qa