The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted source code from user-provided repositories, which serves as an attack surface for indirect prompt injection.
Ingestion points: Source code files are read by the analyze_complexity.py script and provided to the agent's context.
Boundary markers: The skill does not implement explicit boundary markers or isolation for untrusted code content within the analysis workflow.
Capability inventory: The agent can modify local files and execute shell commands (via git), providing a vector for malicious instructions to manifest as harmful code changes.
Sanitization: The tool lacks sanitization or validation of input file content before the agent reasons about optimization.
[COMMAND_EXECUTION]: The analyze_complexity.py script executes local git commands using subprocess.run to identify changed files. These calls are performed with a list of arguments and shell=False, which follows security best practices to prevent shell injection.

complexity-optimizer