The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection by accepting and processing untrusted user data.
Ingestion points: User-provided caption text is accepted via the content parameter in the create_post tool defined in SKILL.md.
Boundary markers: The skill does not implement delimiters or specific instructions to the agent to disregard any instructions that might be embedded within the user-provided caption text.
Capability inventory: The skill is capable of performing network operations to communicate with the Publora MCP server at mcp.publora.com and the REST API at api.publora.com.
Sanitization: There are no explicit sanitization or escaping mechanisms described for handling the content string before it is interpolated into the API requests.

instagram-post