The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill's primary function involves reading and analyzing untrusted data from target codebases, including README files, code comments, and project metadata. This data is then used to generate documentation, providing a surface for indirect prompt injection where an attacker could place malicious instructions inside the analyzed code to manipulate the agent. • Ingestion points: Analysis of codebase files in Step 2 and Step 4 of SKILL.md. • Boundary markers: The skill does not define clear delimiters or specific instructions for the agent to treat file content as untrusted data during analysis. • Capability inventory: The skill performs file system operations, executes shell commands (git, drawio), and invokes other agent skills. • Sanitization: There is no mention of sanitizing or escaping the content extracted from the analyzed codebase.
[COMMAND_EXECUTION]: The skill uses local command execution to perform discovery and artifact processing. • Evidence: In Step 2, it runs 'git branch -a' to gather strategic context. In Step 5, it uses the 'drawio' command-line tool to export architecture diagrams. These operations are scoped to the project directory and are consistent with the skill's stated purpose.

document-service