The Agent Skills Directory

[PROMPT_INJECTION]: The skill analyzes external codebase content, creating a surface for indirect prompt injection where instructions embedded in project files could attempt to influence agent behavior.
Ingestion points: Project files are ingested during indexing via index_project and analyzed via grep_search and read_file.
Boundary markers: Absent; the instructions do not specify delimiters or directions to ignore instructions within analyzed files.
Capability inventory: The skill leverages Bash, Read, Grep, and Glob tools for analysis and fallback operations.
Sanitization: Absent; no validation or filtering of ingested code content is performed.

ln-020-codegraph