The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to use the grep utility to search and filter content within the .understand-anything/knowledge-graph.json file. This is a standard practice for efficient data retrieval in development environments.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from an external file (.understand-anything/knowledge-graph.json) to generate documentation.
Ingestion points: Content is read from .understand-anything/knowledge-graph.json (SKILL.md).
Boundary markers: None are present; the skill assumes the JSON data is well-formatted and safe to summarize.
Capability inventory: The skill uses file reading (grep/read) and file writing (saving to docs/ONBOARDING.md).
Sanitization: There is no explicit sanitization of the strings found in the JSON (e.g., node summaries or tags) before they are placed into the markdown guide.

understand-onboard