From 5385f8b10ed22beb356a8727e6c2b85f12886163 Mon Sep 17 00:00:00 2001
From: Dmitry Pimenov <dmitry@openai.com>
Date: Tue, 25 Mar 2025 16:52:52 -0700
Subject: [PATCH] added readme, fixed typo

---
 examples/mcp/filesystem_example/README.md |  2 +-
 examples/mcp/git_example/README.md        | 25 +++++++++++++++++++++++
 2 files changed, 26 insertions(+), 1 deletion(-)
 create mode 100644 examples/mcp/git_example/README.md

diff --git a/examples/mcp/filesystem_example/README.md b/examples/mcp/filesystem_example/README.md
index 682afc8..1b6b183 100644
--- a/examples/mcp/filesystem_example/README.md
+++ b/examples/mcp/filesystem_example/README.md
@@ -1,6 +1,6 @@
 # MCP Filesystem Example
 
-This example uses the [fileystem MCP server](https://github.com/modelcontextprotocol/servers/tree/main/src/filesystem), running locally via `npx`.
+This example uses the [filesystem MCP server](https://github.com/modelcontextprotocol/servers/tree/main/src/filesystem), running locally via `npx`.
 
 Run it via:
 
diff --git a/examples/mcp/git_example/README.md b/examples/mcp/git_example/README.md
new file mode 100644
index 0000000..35ff7cd
--- /dev/null
+++ b/examples/mcp/git_example/README.md
@@ -0,0 +1,25 @@
+# MCP Git Example
+
+This example uses the [git MCP server](https://github.com/modelcontextprotocol/servers/tree/main/src/git), running locally via `uvx`.
+
+Run it via:
+
+```
+uv run python examples/mcp/git_example/main.py
+```
+
+## Details
+
+The example uses the `MCPServerStdio` class from `agents`, with the command:
+
+```bash
+uvx mcp-server-git
+```
+Prior to running the agent, the user is prompted to provide a local directory path to their git repo. Using that, the Agent can invoke Git MCP tools like `git_log` to inspect the git commit log.
+
+Under the hood:
+
+1. The server is spun up in a subprocess, and exposes a bunch of tools like `git_log()`
+2. We add the server instance to the Agent via `mcp_agents`.
+3. Each time the agent runs, we call out to the MCP server to fetch the list of tools via `server.list_tools()`. The result is cached. 
+4. If the LLM chooses to use an MCP tool, we call the MCP server to run the tool via `server.run_tool()`.