The Agent Skills Directory

[COMMAND_EXECUTION]: The scripts commit_linter.py and generate_changelog.py interact with the local git repository using subprocess.run. These executions are securely implemented using argument lists to avoid shell injection vulnerabilities.
[INDIRECT_PROMPT_INJECTION]: The tool ingests commit messages from git history or user input, which constitutes a vulnerability surface for indirect prompt injection. * Ingestion points: The scripts load commit data from git log output, stdin, and local files. * Boundary markers: The skill does not use explicit boundary markers or delimiters in the generated markdown output to isolate potentially untrusted commit content. * Capability inventory: The skill can execute git commands and update the local filesystem. * Sanitization: The tool validates commit subjects against regex patterns to enforce conventional formats, though the actual message summary is retained in the output.

changelog-generator