The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes search commands like grep to find nodes in a local JSON file based on user-provided arguments. While this is a common method for information retrieval, it represents a potential command argument injection surface.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from the local filesystem and a knowledge graph, which are external inputs that could contain instructions. 1. Ingestion points: .understand-anything/knowledge-graph.json and source files. 2. Boundary markers: None present to separate data from instructions. 3. Capability inventory: Reading local files and executing search tools. 4. Sanitization: No sanitization is applied to the content of the files being explained.

understand-explain