Composio powers 1000+ toolkits, tool search, context management, authentication, and a sandboxed workbench to help you build AI agents that turn intent into action.
TypeScript agentic-aiagentsaiai-agentsaiagentsdeveloper-toolsfunction-callinggpt-4javascriptjsllmllmopsmcppythonremote-mcp-serverssetypescript
docs: add execution flow diagram showing toolkit tools downstream
Sushmithamallesh497861d4a394
diff --git a/docs/content/docs/how-composio-works.mdx b/docs/content/docs/how-composio-works.mdxindex 202016637..1cd0295d1 100644--- a/docs/content/docs/how-composio-works.mdx+++ b/docs/content/docs/how-composio-works.mdx@@ -67,6 +67,26 @@ graph LRMeta tool calls within a session share context through a `session_id`. The agent can search for a tool in one call and execute it in the next without losing state. Tools can also store information (IDs, relationships) in memory for subsequent calls.+### How they work together++The meta tools are how the agent reaches the actual toolkit tools:++```mermaid+graph TD+ A["Agent: 'Create a GitHub issue for this bug'"] --> B["COMPOSIO_SEARCH_TOOLS"]+ B --> C["GITHUB_CREATE_ISSUE"]+ B --> C2["GITHUB_LIST_ISSUES"]+ B --> C3["GITHUB_GET_REPO"]+ B -.- B1["Returns matching tools with schemas"]:::annotation+ C --> D["COMPOSIO_MULTI_EXECUTE_TOOL"]+ D --> E["GitHub API"]+ D -.- D1["Injects user's OAuth token, makes the call"]:::annotation++ classDef annotation stroke-dasharray: 5 5+```++`SEARCH_TOOLS` discovers the right toolkit tools for the task. `MULTI_EXECUTE_TOOL` runs them against the external API with the user's credentials. If the user isn't authenticated yet, `MANAGE_CONNECTIONS` handles that in between.+For large responses or bulk operations (labeling hundreds of emails, processing CSVs), the agent uses `COMPOSIO_REMOTE_WORKBENCH` to run Python with helper functions like `invoke_llm` and `run_composio_tool`.<Card icon={<BookOpen />} title="Tools and toolkits" href="/docs/tools-and-toolkits" description="Meta tools, context management, and execution" />