The Agent Skills Directory

[SAFE]: The skill operates as intended for its stated purpose as a RevOps diagnostic assistant. It follows standard patterns for AI agent skills interacting with an external specialized API.
[COMMAND_EXECUTION]: The skill includes a Python script (scripts/run.py) designed to act as a client for the AI Skills platform. The script's operations are transparent and limited to executing authorized network requests to the vendor's infrastructure.
[EXTERNAL_DOWNLOADS]: The runner script communicates with the official API endpoint at https://ai-skills.ai. This communication is necessary for the skill's core functionality and uses standard HTTPS with support for certificate verification via the certifi package.
[PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it is designed to ingest and process untrusted data from external sources:
- Ingestion points: Processes data from user-provided URLs (materialUrl), file uploads (materialFile), and pasted text (materialText) as defined in references/form-schema.json.
- Boundary markers: No specific delimiters or "ignore embedded instructions" warnings were found in the provided configuration.
- Capability inventory: Local capabilities are limited to outbound HTTP POST requests to the platform API via scripts/run.py.
- Sanitization: No explicit input validation or sanitization logic is present in the client-side script, delegating safety to the platform's backend processing.

ai-revops