The Agent Skills Directory

[COMMAND_EXECUTION]: The skill mentions an installation command npx skills add sales-skills/sales --skills sales-do. This command references a package belonging to the same author ('sales-skills') and is part of the expected functionality for discovering related platform tools.
[DATA_EXPOSURE]: Technical documentation within the skill includes examples for using REST API keys via curl. These examples use standard placeholders like YOUR_KEY and provide security best practices such as the principle of least privilege.
[DATA_EXFILTRATION]: The skill documents the use of 'Connected Content' which involves making HTTP requests to external APIs at message send time. This is a standard feature of the Braze platform used for dynamic personalization and is described here for educational purposes.
[PROMPT_INJECTION]: The skill implements a self-improving mechanism where it reads from and appends to references/learnings.md. While this creates an ingestion point for accumulated knowledge, the operations are limited to the skill's own local reference files.
Ingestion points: Reads accumulated knowledge from references/learnings.md at the start of the interaction.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are defined for the content read from the learnings file.
Capability inventory: The skill provides informational responses and routes to other sub-skills. It does not perform arbitrary code execution or unauthorized network operations.
Sanitization: The skill does not explicitly describe sanitization or validation of the content stored in the learnings file.

sales-braze