Connect LaunchDarkly to Gemini CLI
Give Gemini CLI governed access to LaunchDarkly’s 141 tools through Runlayer, with identity, policy, scanning, and audit on every call.
Set up
Add LaunchDarkly to Gemini CLI in four steps
Admins approve the server once. Everyone in Gemini CLI then gets it through the same governed connection.
- 1
Approve LaunchDarkly in Runlayer
Add LaunchDarkly from the catalog and choose who can use it and which of its tools are allowed.
- 2
Connect Gemini CLI to Runlayer
Users add the Runlayer Plugin to Gemini CLI by following Runlayer’s setup guide. Gemini CLI reaches LaunchDarkly’s 141 tools through the Runlayer Plugin’s search and execute tools instead of listing each one.
Gemini CLI setup guide - 3
Sign in to LaunchDarkly
An admin adds LaunchDarkly credentials to Runlayer once. They never sit in the client’s configuration.
- 4
Use LaunchDarkly from Gemini CLI
LaunchDarkly’s approved tools appear in Gemini CLI. Every call is checked against policy and recorded.
Tools
141 LaunchDarkly tools in Gemini CLI
Runlayer policy decides which of these Gemini CLI users can call.
Apply agentcontrol config approval
Apply an already-approved AgentControl Config approval request.
Apply approval request
Apply an already-approved flag approval request.
Archive flag
Archive a feature flag (reversible).
Check flag rollout complete
Check whether a feature flag is fully rolled out to a single deterministic variation across all traffic-bearing environments.
Check removal readiness
Detailed safety check before removing a feature flag.
Clone agentcontrol config variation
Clone an existing AgentControl Config variation with selective overrides.
Copy flag config
Copy a flag's targeting configuration from one environment to another.
Create agent graph
Create a new agent graph in a project.
Control
What Runlayer can see and stop in Gemini CLI
Some controls apply to every client at the gateway. Others depend on what Gemini CLI exposes to Runlayer.
Gateway policy (available)
Every LaunchDarkly call from Gemini CLI passes through Runlayer, where your policies can allow, alert, mask, or block it.
Audit trail (available)
Each call is recorded with the user, Gemini CLI as the client, the tool, and the policy and scanner decisions.
Client hooks (available)
Capture supported prompts, reasoning, local and MCP tool calls, responses, and scanner decisions in real time.
Stop a call before it runs (available)
Runlayer can deny a tool call before Gemini CLI runs it.
Block output after it runs (available)
Runlayer can block a tool’s output before Gemini CLI’s model sees it.
Redact output in place (not available)
Gemini CLI can only add text to a tool result, not replace it, so Runlayer blocks output it would otherwise mask. The gateway still applies policy to the call.
In Gemini CLI
- Enforcement needs Gemini CLI 0.26.0 or newer.
- When Runlayer blocks a call or its output, Gemini CLI shows the model the block reason instead.
Frequently asked questions
Connect more
