The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill configuration requires the activation of the Rube MCP server located at https://rube.app/mcp. This acts as an external tool dependency for the skill's operations.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from Basecamp (such as message board content and to-do descriptions) and has capabilities to perform write operations or modify project access. This creates a surface for indirect prompt injection where malicious instructions embedded in project content could influence agent behavior.\n
Ingestion points: The skill utilizes tools like BASECAMP_GET_MESSAGE and BASECAMP_GET_BUCKETS_TODOLISTS_TODOS to fetch content from the Basecamp environment as described in SKILL.md.\n
Boundary markers: No specific delimiters or instructions to ignore embedded commands are specified in the provided workflows.\n
Capability inventory: The skill possesses capabilities to write data (e.g., BASECAMP_CREATE_MESSAGE, BASECAMP_POST_BUCKETS_TODOLISTS_TODOS) and manage user permissions (BASECAMP_PUT_PROJECTS_PEOPLE_USERS).\n
Sanitization: The instructions do not define any sanitization or validation protocols for the external content retrieved from Basecamp.

basecamp-automation