Configure and operate Gini runtime state through /api/* endpoints and registered tools.
npx skills add https://github.com/Lilac-Labs/gini-agent --skill gini
Please help me install this Agent Skill. Skill: gini Source: https://github.com/Lilac-Labs/gini-agent/tree/main/skills/agents/gini Command: npx skills add https://github.com/Lilac-Labs/gini-agent --skill gini
This Skill helps users configure, extend, and operate Gini’s runtime correctly by grounding self-directed changes in the exact /api/* endpoints and registered tool behaviors instead of misleading CLI habits.
/api/* and the registered tool catalog, with the CLI treated as a human-operator mirror rather than an agent mechanism.strict, auto, yolo) affects high-risk tool execution so users can prompt accurately without false denials.Ask the agent to explain the correct /api/* call and tool path for a specific capability you want to configure or operate (for example, how to set up a Telegram bridge or connect the managed browser profile).
High-intent search queries and answers about installing and using this skill.
Browser automation and sign-in persistence are configured by guiding the agent through specific /api/* endpoints and registered tools, rather than relying on CLI commands. The CLI acts as a human mirror, while /api/* handles the actual runtime state for browser connectivity cycles.
Setting up a Telegram messaging bridge involves using the /api/* endpoints to register the bridge within the durable runtime state. The skill provides operational playbooks that detail the exact API calls and tool paths required for configuring messaging bridges and recovering from health errors.
Approval modes (strict, auto, yolo) manage high-risk tool execution by defining risk classifications and approval seams. Strict requires explicit approval, auto handles standard risks, and yolo bypasses approvals, ensuring users can prompt accurately without triggering false denials during runtime operations.
Managed-mode browser launch failures and missing Chromium errors occur when the runtime state is misconfigured. The skill provides targeted troubleshooting steps to recover from these common runtime failures by verifying the correct /api/* configurations and browser profile connections.
Yes, MCP server registration and delegated subagents are configured through /api/* recipes. The skill defines the operational instructions for extending the runtime with connectors and subagents, grounding these self-directed changes in the registered tool catalog rather than CLI habits.
The /api/* endpoints and registered tool catalog serve as the primary mechanism for agent runtime configuration, while the CLI is treated as a human-operator mirror. Using /api/* ensures durable runtime state changes for scheduled jobs, connectors, and browser persistence instead of misleading CLI habits.
openyakCustomize Claude Cowork plugin templates, connectors, and workflow rules.
andymaiDiagnose red or stuck GitHub Actions CI runs for the brepjs repository.
vuminhieuDiscover, analyze, and execute MCP capabilities across configured servers.
devtintDetect and exploit NoSQL injection vulnerabilities in MongoDB, CouchDB, and Redis APIs.