The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes shell commands to perform its tasks, specifically git ls-files to retrieve a list of tracked files and the open command to display the final HTML output in the user's browser. These actions are within the expected scope of a documentation generator.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it is designed to ingest and analyze untrusted data from a local codebase. An attacker could place malicious instructions inside the files being analyzed (e.g., in a README or code comment) to influence the agent's behavior during the documentation process.
Ingestion points: Files read from the sourceRoot during the 'Investigation Workflow' (Step 2).
Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions when the agent processes file contents.
Capability inventory: Filesystem reading, directory walking, file writing to the docs/ folder, and execution of git and open commands.
Sanitization: The skill lacks explicit instructions for sanitizing or escaping content extracted from files before it is processed by the AI.

visual-docs