The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill demonstrates a vulnerability surface for indirect prompt injection by ingesting and processing data from external sources.
Ingestion points: The skill retrieves data from business websites and the MentionedOn service via the get_website_fixes and inspect_business_website tools, as described in SKILL.md.
Boundary markers: Absent. The instructions lack specifications for using delimiters or warnings to ignore potentially malicious instructions embedded within the fetched external data.
Capability inventory: The agent possesses tools such as draft_website_fixes and verify_website_fixes that are used to process, store, and verify content based on the ingested data.
Sanitization: Absent. There are no requirements or instructions for the agent to sanitize, validate, or filter the external content before it is processed into drafts or used for verification.

mentionedon-website-fixes