The Agent Skills Directory

[SAFE]: The skill facilitates interaction with the official Sinch CLI (@sinch/cli), a legitimate developer tool for managing cloud communication services. All external resources, including documentation links (developers.sinch.com) and packages, originate from the verified vendor infrastructure.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an ingestion surface through local configuration files like sinch.json and .env. These files are used to interpolate parameters into shell commands. However, the skill mitigates risk by instructing the agent to verify command syntax using sinch <cmd> --help and focusing on well-defined CLI parameters, which is standard behavior for development tools.
[CREDENTIALS_UNSAFE]: The skill correctly identifies the risk of hardcoded secrets and explicitly instructs the user and agent to use the sinch secrets command to store sensitive data in the OS keychain rather than in source code or environment files.

sinch-cli