The Agent Skills Directory

[SAFE]: The skill serves as a legitimate configuration and workflow guide for the Reply toolkit via Composio. It utilizes documented tool patterns and points to official resources.
[PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection as it processes dynamic tool schemas and execution results to inform subsequent agent actions.
Ingestion points: Data returned from RUBE_SEARCH_TOOLS and the output of various Reply tools integrated via MCP (SKILL.md).
Boundary markers: No explicit delimiters or instructions to ignore instructions embedded in tool outputs are provided.
Capability inventory: The skill utilizes RUBE_MULTI_EXECUTE_TOOL and RUBE_REMOTE_WORKBENCH for executing tasks and managing remote environments (SKILL.md).
Sanitization: No explicit sanitization or validation logic for external data is defined in the instruction set.

reply-automation