Skip to content
docs

LangChain and LangGraph and Deedglass

Give a LangChain agent Deedglass’s tools in one file. The file starts Deedglass’s MCP server (npx -y deedglass mcp serve) and hands its tools to the agent: each call goes through your policy, the risky ones wait for a person, and each one is sealed as a deed you can check later.

Tested with langchain 1.4.6 with the mcp extra and langchain-anthropic 1.7.6 (2026-10-21).

Before you start

Section titled “Before you start”
  1. In any folder, run npx deedglass init (Docker, about 15 seconds; the quickstart). It starts Deedglass and a demo Postgres with 500 orders, and saves an agent key in .deedglass/quickstart.key.
  2. Run the file below from that folder (or a folder inside it). deedglass mcp serve looks for the key there; from anywhere else, set DEEDGLASS_API_KEY and DEEDGLASS_SERVER yourself (how mcp serve finds them).
  3. Under the postgres-safe policy that init turns on, reads run, DROP and TRUNCATE are refused, and a write waits for a person in Console › Approvals before it runs.

The MCP route

Section titled “The MCP route”
examples/frameworks/langchain/agent.py
# Run where you ran `npx deedglass init`. Needs ANTHROPIC_API_KEY (langchain-anthropic).
import asyncio
from langchain.agents import create_agent
from langchain.mcp import MCPAdapter


CONFIG = {"mcpServers": {"deedglass": {"command": "npx", "args": ["-y", "deedglass", "mcp", "serve"]}}}


async def main():
    async with MCPAdapter(CONFIG) as adapter:
        tools = await adapter.list_tools()
        agent = create_agent("claude-sonnet-5", tools)
        print((await agent.ainvoke({"messages": [{"role": "user", "content": "How many rows are in public.orders?"}]}))["messages"][-1].content)


asyncio.run(main())

Install and run:

Terminal window
pip install "langchain[mcp]==1.4.6" langchain-anthropic==1.7.6
python agent.py

Model key: ANTHROPIC_API_KEY (the model string is the one in LangChain’s MCP page). create_agent takes any chat model LangChain supports.

Tool names: Deedglass lists its tools with names every model API takes: postgres_query, postgres_execute, deedglass_check_approval (letters, digits, _ and -; Anthropic’s rule is ^[a-zA-Z0-9_-]{1,128}$ and OpenAI’s allows the same characters up to 64). Your policies and the sealed records keep the dotted name, postgres.query; Deedglass maps one to the other (MCP.md). Deedglass 0.7 and earlier listed the dotted names, which these APIs refuse. langchain.mcp is marked beta in 1.4.6 (it warns on import).

What you see: the answer (500 rows), and in Console › Ledger or npx deedglass actions list one sealed postgres.query action per call, with the rule that allowed it. If the agent tries a write, the tool answers that it is waiting for approval and nothing has run; once someone approves, deedglass_check_approval lets the agent carry on. A refused call answers with the policy’s sentence and a record id.

First start: npx -y deedglass can take several seconds the first time. If a framework’s MCP connect timeout is shorter, run npx -y deedglass version once first, or install the CLI (INSTALL.md) and use deedglass as the command.

The native adapter

Section titled “The native adapter”

The Python SDK has a LangChain adapter, DeedglassMiddleware, built on LangChain’s wrap_tool_call middleware: it asks Deedglass before each of your own tools runs and reports the result after (SDK_PYTHON.md). For LangGraph’s ToolNode, govern_tool wraps tools directly with pre-execution policy checks and clean error reporting. It isn’t on PyPI yet. For LangChain.js, @deedglass/sdk/langchain has governTool and deedglassTools (its page).

Check it without a model key

Section titled “Check it without a model key”

examples/frameworks/langchain/ has the file, its pinned packages and a check file that needs no model key: it starts deedglass mcp serve through the same framework client, and runs postgres_query once with ainvoke(). From the folder where init ran:

Terminal window
path/to/deedglass/examples/frameworks/run.sh langchain            # install the pins in a temp folder, run the check
path/to/deedglass/examples/frameworks/run.sh langchain --agent    # also run the agent file when its model key is set

The workflow frameworks.yml runs every check against a fresh init stack each week and on pull requests that touch these files. It isn’t a required check: a framework release that doesn’t install is reported as skipped, not as a failure.

Sources and what isn’t verified

Section titled “Sources and what isn’t verified”

Every framework API in the files above comes from the framework’s own docs, checked 2026-10-21:

Not verified:

  • Renaming a tool by setting .name: LangChain documents name as a tool field, but not that the MCP adapter keeps calling the server’s name; we observed it in 1.4.6.
  • A run with a live model (no model key in CI).