The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses shell commands, specifically git log, git ls-files, and git diff, to analyze repository history. These commands rely on file paths and timestamps extracted directly from documentation files within the project, which could be manipulated if documentation is compromised.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection due to its processing of untrusted local data.
Ingestion points: The skill reads specific markdown sections, such as ## Structure and ## Implementation, from files located in the docs/ directory. It also processes the output of git commit messages and stats.
Boundary markers: There are no explicit instructions or delimiters defined to prevent the agent from accidentally executing instructions found within the parsed documentation sections.
Capability inventory: The skill allows for the execution of git-related subprocess calls and read operations on local files.
Sanitization: No sanitization or validation logic is present to ensure that the content read from documentation files or git history is purely data and does not contain malicious instructions.

validate-docs