The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to fetch live documentation from external websites like developer.woocommerce.com and developer.wordpress.org using the WebSearch and WebFetch tools. This external content is intended to influence subsequent actions such as writing test code or executing shell commands via the Bash tool. The lack of explicit boundary markers or sanitization for this external data creates a potential surface for indirect prompt injection, where malicious content on a web page could attempt to influence the agent's behavior.
Ingestion points: Documentation fetched from external URLs via WebSearch and WebFetch (referenced in the 'Before writing code' section).
Boundary markers: No specific delimiters or instructions to ignore embedded commands within the fetched documentation are provided.
Capability inventory: The skill allows file manipulation (Write, Edit) and shell command execution (Bash) which could be misused if the agent obeys malicious instructions embedded in fetched documentation.
Sanitization: No explicit sanitization or validation of the fetched documentation is mentioned.

woo-testing