The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes raw HTML content from the article detail API, which could contain malicious instructions designed to influence the agent's behavior.
Ingestion points: The content field returned by the article/detail action in SKILL.md is specified as raw HTML and displayed to the user.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to ignore potentially malicious content embedded within the fetched article text.
Capability inventory: The skill has permissions to perform network operations (POST requests) to a Supabase-hosted API endpoint (swwgusxefvtbdwyoppir.supabase.co).
Sanitization: The skill explicitly prohibits HTML escaping or secondary encoding of the content field, increasing the likelihood that embedded scripts or instructions could be processed by the agent or the user's browser.

gigglepro-article