The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill provides an interface to compute clusters from data stored in external storage buckets, which can serve as a vector for indirect prompt injection if the bucket content is attacker-controlled.
Ingestion points: The /ai/clusters endpoint (and corresponding client.ai().clusters().compute() method) ingests data from an "embedded storage bucket".
Boundary markers: No explicit boundary markers or "ignore instructions" delimiters are provided in the SDK examples for processing this external data.
Capability inventory: The skill possesses extensive capabilities to modify infrastructure, including creating and deleting private networks, WireGuard VPN gateways, and virtual cross-connects.
Sanitization: The skill documentation does not mention sanitization or validation of the content retrieved from the storage buckets before processing.
[DATA_EXPOSURE_AND_EXFILTRATION]: The skill handles sensitive cryptographic credentials required for networking security.
The API response schema for WireGuard Peers (/wireguard_peers) includes the private_key field, which is a high-value secret.
The skill documentation correctly recommends using TelnyxOkHttpClient.fromEnv() to load API credentials from environment variables, which is a standard security best practice.

telnyx-networking-java