The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes content from external research papers fetched via the paper-navigator dependency to generate research ideas and trees.
Ingestion points: External research paper metadata and full text retrieved via paper-navigator scripts, and local memory files located at /memory/ideation-memory.md.
Boundary markers: There are no instructions to delimit or specifically handle potentially malicious instructions embedded in the external research content.
Capability inventory: The skill has access to powerful tools including write_file, edit_file, read_file, think_tool, and execute.
Sanitization: The skill lacks sanitization or validation routines for data processed from external sources before it is used in subsequent reasoning steps.
[NO_CODE]: This skill contains only markdown files and does not ship with any executable scripts, binary files, or source code.

research-ideation