The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read, analyze, and modify PHP source code within a user's project, creating a surface for indirect prompt injection where malicious instructions could be embedded in the code being processed.
Ingestion points: Project files are accessed using the Read, Glob, and Grep tools as part of the mapping and implementation workflow.
Capability inventory: The skill requires the ability to modify files (Write, Edit) and execute shell commands (Bash) for validation and debugging.
Boundary markers: No explicit instructions are provided to the agent to distinguish between implementation code and potentially malicious instructions within the source files.
Sanitization: The skill does not define methods for sanitizing or filtering the content of the ingested project files before processing.

api-platform-dto-resources