The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill references and fetches tool schemas and configuration from rube.app and composio.dev. These are official endpoints for the integration and no malicious use was detected.\n- [PROMPT_INJECTION]: The skill uses RUBE_SEARCH_TOOLS to retrieve tool metadata and execution plans at runtime, which constitutes an indirect prompt injection surface.\n
Ingestion points: External tool metadata and recommendations retrieved from the RUBE_SEARCH_TOOLS endpoint.\n
Boundary markers: Absent in the skill instructions.\n
Capability inventory: Connection management and multi-tool execution capabilities via RUBE_MULTI_EXECUTE_TOOL and RUBE_REMOTE_WORKBENCH.\n
Sanitization: The instructions do not define sanitization or validation steps for the retrieved schemas or plans before execution.

ravenseotools-automation