The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze and summarize external, untrusted content from a local codebase, which is a common requirement for documentation generation tasks.\n- Ingestion points: The agent reads all source files, READMEs, and git history (using git log) during the 'Deep Reading' phase to construct the narrative guide.\n- Boundary markers: The skill does not explicitly instruct the agent to use delimiters or warnings to isolate processed code content, though the use of a predefined output template provides structural guidance.\n- Capability inventory: The skill is limited to file system read and write operations and standard version control commands within the project directory. It does not perform network operations or execute arbitrary code from the analyzed files.\n- Sanitization: No specific sanitization or filtering of codebase comments or documentation is performed before the agent processes them.

literate-guide