The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill instructions involve processing data retrieved from a graph structure, which constitutes an indirect injection surface. * Ingestion points: The skill uses iskron_search and iskron_look to retrieve data from the graph (SKILL.md, Sections 1 and 2). * Boundary markers: There are no explicit instructions to use delimiters or ignore embedded instructions in the processed data. * Capability inventory: The skill utilizes iskron_update and iskron_arrow to modify nodes and relationships within the graph (SKILL.md, Section 2). * Sanitization: No sanitization or filtering of external content is described.
[SAFE]: All functional elements, including the iskron_* tools and the cases.md reference, are consistent with the author's internal ecosystem. No unauthorized network activity, credential theft, or persistence mechanisms were found.

inquiry