The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The code_quality_analyzer.py script processes untrusted codebase files to perform security and quality audits. This creates a surface for indirect prompt injection if malicious instructions are embedded within the analyzed files.
Ingestion points: The script walks through user-specified directories and reads the contents of all source code and configuration files.
Capability inventory: The script performs local file system reads and provides summary reports; it does not have network access or the ability to modify system configuration.
Boundary markers: The script does not wrap code snippets or findings in explicit boundary markers or instructions to ignore embedded commands.
Sanitization: The tool identifies patterns using regex but does not sanitize the content before presenting it in the output.
[COMMAND_EXECUTION]: The skill package includes and utilizes local Python scripts for project generation and auditing. Workflows in SKILL.md explicitly direct the agent to execute scripts/project_scaffolder.py and scripts/code_quality_analyzer.py. These operations are essential to the skill's primary function and are restricted to the local scripts provided within the skill environment.

senior-fullstack