The Agent Skills Directory

[COMMAND_EXECUTION]: The skill is designed to wrap the lightpanda command-line utility. It provides instructions on how the agent should execute this binary to perform tasks like fetch, serve, and mcp. This assumes the binary is already present on the host system's PATH.
[INDIRECT_PROMPT_INJECTION]: The skill's primary function is to fetch content from arbitrary external URLs and return that data (in HTML, Markdown, or Semantic Tree formats) to the agent's context. This creates a surface for indirect prompt injection if the fetched content contains malicious instructions designed to influence the agent's behavior.
Ingestion points: Untrusted data enters the agent context through the lightpanda fetch command via the URL parameter and the resulting page content (SKILL.md).
Boundary markers: The instructions do not specify the use of delimiters or specific system instructions to ignore potential commands within the fetched content.
Capability inventory: The skill has the capability to execute the local lightpanda binary and perform network requests to any reachable URL.
Sanitization: There is no evidence of sanitization or filtering of the fetched content beyond the built-in formatting options (like --strip_mode).

lightpanda