The Agent Skills Directory

[COMMAND_EXECUTION]: The file tests/test_scanner.py uses the os.system function with f-string interpolation to initialize and commit to temporary Git repositories for testing purposes. This pattern is susceptible to command injection if the generated directory paths contain shell-sensitive characters.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from Git commit history (messages and file contents) and incorporates it into generated Markdown, HTML, and PDF reports without sanitization or boundary markers. This creates an attack surface for indirect prompt injection where a repository contributor could attempt to influence the agent's behavior during report interpretation. Evidence: Ingestion occurs in src/analyzers/base_analyzer.py; report generation in src/reporters/markdown_reporter.py and src/reporters/html_reporter.py lacks escaping or delimiters for external strings.
[SAFE]: The skill documentation includes extensive privacy and data security declarations, confirming that all analysis is performed locally and no data is transmitted to external servers.

code-analysis