The Agent Skills Directory

[SAFE]: The skill instructs the agent to interact with official LaunchDarkly API endpoints at https://app.launchdarkly.com/api/v2 for configuration management. All network operations are directed at the vendor's own services and follow standard API integration practices.- [SAFE]: The skill provides guidance for the agent to locate API credentials in standard environment variables (e.g., LAUNCHDARKLY_API_KEY) or the Claude desktop configuration file (~/.claude/config.json). This access is a functional requirement for managing the vendor's service and does not involve unauthorized data exfiltration.- [SAFE]: The skill references official LaunchDarkly SDKs and documentation repositories on GitHub. These references are for educational and configuration purposes and do not involve remote code execution from untrusted sources.- [NO_CODE]: The skill does not include any executable scripts, binaries, or automated tools. It consists entirely of Markdown documentation providing instructions and code snippets for developer reference.

online-evals