The Agent Skills Directory

[COMMAND_EXECUTION] (SAFE): The skill executes the speakeasy status command. This is the intended primary function for retrieving workspace metadata and does not involve arbitrary or unsanitized command execution.- [DATA_EXPOSURE] (SAFE): The skill mentions the use of SPEAKEASY_API_KEY and the configuration file at ~/.speakeasy/config.yaml. Accessing these is required for the tool to function and authenticate with the Speakeasy service, and no evidence of exfiltration to third parties was found.- [INDIRECT_PROMPT_INJECTION] (LOW): The skill has a surface for indirect prompt injection as it processes output from the speakeasy status command which might contain data from external sources (like repository names or versions). * Ingestion points: Data enters the agent context through the output of the speakeasy status --output json command. * Boundary markers: No specific delimiters are defined in the instructions to separate tool output from instructions. * Capability inventory: The skill allows command execution specifically via the speakeasy CLI. * Sanitization: No explicit sanitization of the tool output is performed before it is presented to the agent.

check-workspace-status