The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the user or agent to install the @membranehq/cli package from the official npm registry. This is a legitimate installation of the vendor's command-line tool required for the skill's operation.
[COMMAND_EXECUTION]: The instructions involve executing multiple membrane CLI commands to perform tasks such as authentication, searching for connectors, and running compliance actions. These executions are consistent with the skill's intended purpose of managing TruNarrative data.
[PROMPT_INJECTION]: The skill includes an attack surface for indirect prompt injection via user-supplied parameters.
Ingestion points: User-provided intent strings and JSON input objects in SKILL.md used for action discovery and execution.
Boundary markers: Absent; the instructions do not specify delimiters or constraints for the agent when handling these inputs.
Capability inventory: The agent utilizes shell commands through the membrane CLI to interact with external APIs.
Sanitization: Absent; the skill does not explicitly provide validation or escaping logic for user input before it is interpolated into shell commands.

trunarrative