The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a shell command (find) to scan the project folder and generate a file map. While this is core to its functionality, the command uses a path variable ([project-folder]) that can be influenced by user input or environment context.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface (Category 8) because it automatically reads the first lines of text files in the project directory to generate descriptions.
Ingestion points: Step 3 of SKILL.md describes reading the first line of .md, .txt, and .csv files.
Boundary markers: None identified; content is interpolated into the generated context.md file.
Capability inventory: Uses find and read operations via the agent's shell tools to populate documentation.
Sanitization: No explicit sanitization or instruction-ignoring delimiters are used when processing file headers.

project-onboarding