The Agent Skills Directory

[SAFE]: No malicious patterns or security vulnerabilities were identified. The skill follows best practices for authentication and uses official vendor tools.
[EXTERNAL_DOWNLOADS]: Installs the @membranehq/cli package from npm, which is the official tool provided by the authoring organization.
[COMMAND_EXECUTION]: Instructs the agent to use the membrane CLI to perform integration tasks, search for actions, and run API requests.
[PROMPT_INJECTION]: The skill provides an interface to retrieve data from LeadBoxer, creating a surface for potential indirect prompt injection.
Ingestion points: Lead details, events, and segments are fetched from the LeadBoxer API (SKILL.md).
Boundary markers: None explicitly provided in the skill instructions.
Capability inventory: Uses the membrane CLI for network requests and data updates (SKILL.md).
Sanitization: Relies on the platform's action schema validation.

leadboxer