The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill includes scripts/generate_report.py, which contains a generate_script function that creates bash scripts from a template and explicitly sets executable permissions using chmod 0o755. This allows the agent to generate and potentially execute new shell scripts at runtime, which is a significant security risk if the script content is derived from untrusted inputs.
[INDIRECT_PROMPT_INJECTION]: The profiling logic in scripts/analyze_results.py scans arbitrary file systems, making the skill vulnerable to indirect prompt injection through maliciously crafted file names or directory structures.
Ingestion points: The analyze_directory method in scripts/analyze_results.py uses rglob("*") to recursively traverse and process all files and directories in a target path provided by the user.
Boundary markers: The skill does not implement delimiters or explicit instructions to treat discovered file metadata as untrusted data, allowing it to flow directly into reports or subsequent agent context.
Capability inventory: The skill frontmatter permits Bash(cmd:*), and the generate_report.py utility provides a mechanism to write and execute scripts.
Sanitization: There is no evidence of sanitization or escaping for file paths, names, or metadata before they are used in report generation or as parameters for other script operations.

profiling-application-performance