The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the user to install the @membranehq/cli package from the npm registry. This is an official scoped package provided by the vendor (membranedev) to facilitate interaction with their platform.
[COMMAND_EXECUTION]: The instructions utilize the membrane command-line interface to manage logins, create service connections, and execute automated actions. These commands are standard operations within the platform's ecosystem.
[PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection as it processes untrusted data from emails and document attachments via Parsio.io.
Ingestion points: Untrusted external data enters the agent context from Parsio.io documents and parser outputs.
Boundary markers: The skill does not currently define explicit delimiters or instructions to ignore potential commands embedded within the parsed data.
Capability inventory: The skill can execute actions via the CLI, which may include file system or network operations depending on the specific actions created/run.
Sanitization: There is no evidence of explicit sanitization or filtering of the external data before processing.
[DATA_EXFILTRATION]: The skill demonstrates a positive security posture by explicitly advising against manual API key management and instead utilizing the platform's centralized connection system, reducing the risk of credential exposure.

parsioio