Klavis Strata

Watch this video tutorial for a complete understanding of how Strata works:

1. Discover Server Categories or Actions

discover_server_categories_or_actions- find relevant categories or actions based on user intent. No semantic search!

If detail level is 'categories_only', the details will be a list of category names only. Next step prefer to use get_category_actions tool to get the actions for the categories.

If detail level is 'full_details', the details will be a list of category names with their actions details included. This happens when the server has only a few actions. Next step prefer to use execute_action tool to execute the actions.

If detail level is 'categories_and_actions', the details will be a list of category names and action names. This happens when using external tools. Next step prefer to use get_action_details tool to get the details of the actions.

- user_query(string, required): Natural language user query to filter results.
- server_names(array, required): List of server names to discover categories or actions.

get_category_actions- retrieve all action names within specified categories.

- category_names(array, required): List of categories to get actions for

get_action_details- get full schema and parameters for a specific action.

- category_name(string, required): The name of the category
- action_name(string, required): The name of the action/operation within the category

execute_action- run actions with parameters and get results.

- server_name(string, required): The name of the server
- category_name(string, required): The name of the category to execute the action for
- action_name(string, required): The name of the action/operation to execute
- path_params(string, optional): JSON string containing path parameters for the action
- query_params(string, optional): JSON string containing query parameters for the action
- body_schema(string, optional, default: "{}"): JSON string containing request body for actions
- include_output_fields(array, optional): Optional but strongly recommended when you know the response_schema of this action from previous tool calls: Array of field paths to include in the response. Only these fields will be returned. Use dot notation for nested fields (e.g., "author.displayName").
- maximum_output_characters(integer, optional): Optional: Maximum number of characters to return in the response. If the response exceeds this limit, it will be truncated. Prefer include_output_fields over this.

search_documentation- find relevant information only when needed.

- query(string, required): Search keywords that match API documentation terms. Best practices: (1) Use resource names like 'users', 'projects', 'files', (2) Add actions for precision like 'user create' or 'project delete', (3) Avoid filler words like 'how to', 'show me', 'all the' - focus on the core terms that appear in endpoint names and descriptions.
- server_name(string, required): Name of the server to search within.
- max_results(integer, optional, default: 10, minimum: 1, maximum: 50): Number of results to return. Default: 10

handle_auth_failure- handle authentication only when needed.

Apigene MCP Gateway is the runtime layer that connects AI agents to APIs and MCP servers via Model Context Protocol.

One API key, dozens of capabilities for your AI agent. Zero provider auth.

Intelligent rate limiting for agent-to-agent communication — per-agent quotas, sliding window throttling, burst allowances, and fair-use enforcement across multi-agent MCP deployments.

Unified MCP & skill management gateway with progressive disclosure. Manages multiple MCP servers as Agent Apps, loading tool schemas on demand for 99% context token savings. Shared across Claude Code, Codex, OpenCode and more.

65+ AI tools as an MCP server. Research, write, code, scrape, translate, analyze, agent memory, workflows. Pay per call from $0.006.

The Ultimate Control Plane for MCP Unlock the full power of Model Context Protocol with zero friction. One-Click GPT Integration: Bridge the gap between MCP servers and ChatGPT/LLMs instantly. No more manual config hunting. Pro-Level Orchestration: Manage, monitor, and toggle multiple MCP tools from a single, intuitive dashboard. Secure by Design: Built-in support for complex auth flows and 2FA, making enterprise-grade tool integration seamless. Streamlined Debugging: Test queries and inspect tool responses in real-time without leaving the hub. Stop wrestling with JSON configs. Start building agentic workflows that actually work.

Run MCP tools in production without managing your own server — built-in retries, permissions, and observability.

Magica is your all-in-one AI platform, offering 2500+ cutting-edge tools under a single subscription.

Dynamically search and call tools using UnifAI Network