The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process content from external, untrusted URLs for the purpose of capturing screenshots. This introduces a potential surface for indirect prompt injection if the model attempts to parse or interpret data from the targeted web pages.
Ingestion points: Target URLs processed by the strale_screenshot_url and context_dev_capture_screenshot tools as described in SKILL.md and references/sandbase-api-map.md.
Boundary markers: The skill does not define specific delimiters or instructions to ignore instructions embedded within the target web pages.
Capability inventory: The skill uses sandbase_call_tool to execute the screenshot capture functionality (SKILL.md).
Sanitization: There are no explicit sanitization or validation procedures defined in the skill files for the content retrieved from external URLs.

screenshot-capture