The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill directs users to install the appwrite package, which is the official library provided by the vendor for interacting with their platform.
[CREDENTIALS_UNSAFE]: Code examples appropriately use environment variables (os.environ) to handle sensitive data like the APPWRITE_API_KEY, avoiding the risk of hardcoded secrets.
[PROMPT_INJECTION]: The skill identifies surfaces for indirect injection where external data is processed in web handlers or serverless functions.
Ingestion points: Untrusted data enters the agent context via request.json, request.args, and context.req.body_json in the SSR and Function examples in SKILL.md.
Boundary markers: None are present in the provided code snippets.
Capability inventory: The skill enables database writes (tables_db.create_row), file uploads (storage.create_file), and user management (users.create).
Sanitization: The snippets show basic data handling without explicit sanitization or validation logic.

python