The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill facilitates network communication with vendor-controlled infrastructure and numerous third-party API providers.
It connects to gateway.maton.ai, ctrl.maton.ai, and connect.maton.ai for connection management and request proxying.
It provides documentation and routing paths for approximately 80 external domains belonging to third-party services like Google, Microsoft, Slack, Salesforce, and Stripe.
[COMMAND_EXECUTION]: The skill provides Python code snippets in SKILL.md and reference files designed for the agent to execute.
These snippets use the urllib.request and requests libraries to perform HTTP operations and handle JSON data.
[PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection by design, as its primary function is to ingest data from external sources.
Ingestion points: Data is ingested from external API responses across many services (e.g., Slack messages, Gmail content, GitHub issues) via the Maton gateway proxy.
Boundary markers: The provided documentation and code examples do not include the use of delimiters or specific safety instructions to isolate untrusted data returned from these APIs.
Capability inventory: The skill demonstrates capabilities for network requests (urllib.request, requests) and data parsing (json.load, json.dumps).
Sanitization: There is no evidence of filtering or sanitizing the content returned from third-party APIs before it is incorporated into the agent's context.

api-gateway