The Agent Skills Directory

[Command Execution] (SAFE): The skill utilizes the speakeasy CLI for context retrieval and task execution. These operations are essential to the primary purpose of the skill and do not involve unauthorized system modifications.
[Data Exposure & Exfiltration] (SAFE): The speakeasy agent feedback command is used to report workflow outcomes. While it transmits data to an external service, it is a documented feature of the tool and is not designed to target sensitive files or hardcoded credentials.
[Indirect Prompt Injection] (SAFE): The skill ingests data from the speakeasy agent context command. Although this data is external, there are no instructions to execute untrusted code found within that data, and the surface is restricted to the CLI's own output.

speakeasy-context