The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: Fetches and installs companion feature management skills from the official LaunchDarkly GitHub repository (launchdarkly/ai-tooling) using the npx skills add command to extend agent capabilities.
[COMMAND_EXECUTION]: Executes standard package management commands (e.g., npm install, pip install, go get) to integrate LaunchDarkly SDKs and uses curl or ldcli for interaction with the LaunchDarkly API.
[SAFE]: Implements robust credential management by mandating the use of environment variables for SDK keys and tokens, and explicitly warning against hardcoding sensitive information in source code.
[SAFE]: Enforces human-in-the-loop security via mandatory blocking decision points that require user approval before the agent writes secrets to the filesystem, modifies non-LaunchDarkly project dependencies, or configures local MCP servers.

onboarding