The Agent Skills Directory

[External Download/Remote Code Execution] (HIGH): The skill instructs the user to add https://rube.app/mcp as an MCP server. This domain is not within the trusted repository or organization list. Since MCP servers define the executable tools and logic available to the agent, this grants an untrusted third party significant control over the agent's runtime environment.
[Indirect Prompt Injection] (HIGH): The skill is designed to automate Dailybot tasks, which involves reading external data (messages, standups) that may contain malicious instructions.
Ingestion points: Data returned from Dailybot via Rube MCP tools.
Boundary markers: Absent; there are no instructions to use delimiters or ignore instructions embedded in the processed data.
Capability inventory: The presence of RUBE_MULTI_EXECUTE_TOOL and RUBE_REMOTE_WORKBENCH provides high-privilege write and execution capabilities.
Sanitization: Absent; no mention of validating or escaping content before it is passed to tool arguments.
[Dynamic Execution] (MEDIUM): The skill relies on RUBE_SEARCH_TOOLS to fetch schemas at runtime, which are then used to build tool calls in RUBE_MULTI_EXECUTE_TOOL. This dynamic discovery of execution schemas from an untrusted source increases the risk of the agent being manipulated into executing unintended commands.
[Credential Exposure] (MEDIUM): The setup involves RUBE_MANAGE_CONNECTIONS which directs users to an external authentication link on the untrusted rube.app domain, potentially exposing service tokens or sensitive connection metadata.

dailybot-automation