The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes source code from the user's project to trace prop origins and drilling paths. This creates a surface for indirect prompt injection if malicious instructions are embedded within the codebase being analyzed (e.g., in code comments or large string literals).
Ingestion points: Phased search and read operations on *.ts and *.tsx files within the local project directory (SKILL.md, Phases 1-3).
Boundary markers: No specific delimiters or "ignore instructions" warnings are defined for the file content being processed.
Capability inventory: The skill utilizes file reading (Read), pattern matching (Grep), shell command execution (via Bash compatibility mapping), and external search capabilities (via WebSearch and exa MCP mappings).
Sanitization: The skill does not perform sanitization or validation of the code content before analysis.

prop-drill