The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructions direct the user to connect to an external MCP server endpoint at https://rube.app/mcp. This is the primary infrastructure provided by the vendor (Rootcastle) for the skill's functionality and does not involve direct execution of untrusted scripts.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it is designed to ingest and process untrusted data from Intercom conversations and contact fields.
Ingestion points: External data is ingested via tools such as INTERCOM_GET_CONVERSATION, INTERCOM_SEARCH_CONTACTS, and INTERCOM_LIST_CONVERSATIONS as described in SKILL.md.
Boundary markers: The instructions do not specify the use of clear delimiters or 'ignore' instructions for the agent when processing external conversation bodies.
Capability inventory: The agent has the capability to modify external state by using tools like INTERCOM_REPLY_TO_CONVERSATION, INTERCOM_CLOSE_CONVERSATION, and INTERCOM_DETACH_A_CONTACT.
Sanitization: The skill documentation includes a specific mitigation note advising the agent to sanitize HTML input in conversation bodies to prevent rendering issues.

intercom-automation