The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads and processes untrusted source code from a user's codebase. Maliciously crafted comments or string literals within the code being analyzed could influence the agent's behavior.
Ingestion points: Workflow steps 1 and 3 in SKILL.md describe walking the project tree and inspecting symbols to catalog the codebase.
Boundary markers: No explicit instructions are provided to the agent to ignore or delimit instructions found within the code being analyzed.
Capability inventory: Workflow step 5 describes the capability to insert or update documentation in-place on the filesystem.
Sanitization: There are no mentions of sanitizing or validating the content extracted from the code before using it to generate documentation or writing it back to files.
[NO_CODE]: The skill consists entirely of natural language instructions and examples within a Markdown file. No executable scripts, binaries, or configuration files are provided.

code-documentation