The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted data in the form of PHP code snippets and data flow evidence (EVID) provided by an external pipeline.
Ingestion points: The skill relies on evidence provided by php-audit-pipeline as mentioned in SKILL.md.
Boundary markers: The instructions do not specify the use of delimiters or specific warnings to the agent to disregard instructions potentially embedded within the PHP code evidence being audited.
Capability inventory: While the skill itself is instructional, it is designed to be used by an agent that typically has access to file system and terminal tools; malicious instructions in audited code could attempt to leverage these capabilities.
Sanitization: There are no instructions for sanitizing or validating the input data before it is processed by the agent.

php-file-audit