Skip to content
docs

OpenAI Agents SDK and Deedglass

Give an OpenAI Agents SDK agent, in Python or TypeScript, 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. There is also a native adapter in @deedglass/sdk.

Tested with openai-agents 0.24.0 for Python and @openai/agents 0.21.0 (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/openai-agents/agent.py
# Run where you ran `npx deedglass init`. Needs OPENAI_API_KEY.
import asyncio
from agents import Agent, Runner
from agents.mcp import MCPServerStdio


async def main():
    params = {"command": "npx", "args": ["-y", "deedglass", "mcp", "serve"]}
    async with MCPServerStdio(name="deedglass", params=params) as server:
        agent = Agent(name="orders", instructions="Answer questions about orders.", mcp_servers=[server])
        result = await Runner.run(agent, "How many rows are in public.orders?")
        print(result.final_output)


asyncio.run(main())
examples/frameworks/openai-agents/agent.ts
// Run where you ran `npx deedglass init`. Needs OPENAI_API_KEY.
import { Agent, MCPServerStdio, run } from '@openai/agents';


const server = new MCPServerStdio({ name: 'deedglass', fullCommand: 'npx -y deedglass mcp serve' });
await server.connect();
try {
  const agent = new Agent({ name: 'orders', instructions: 'Answer questions about orders.', mcpServers: [server] });
  const result = await run(agent, 'How many rows are in public.orders?');
  console.log(result.finalOutput);
} finally {
  await server.close();
}

Install and run:

Terminal window
pip install openai-agents==0.24.0 && python agent.py
npm install @openai/agents@0.21.0 zod@4.6.7 && node agent.ts

Model key: OPENAI_API_KEY. The agent uses the SDK’s default model.

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. include_server_in_tool_names (includeServerInToolNames in TypeScript) is no longer needed; turn it on only to tell several servers’ tools apart.

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”

@deedglass/sdk has an OpenAI Agents adapter for TypeScript (its page). deedglassTools asks Deedglass for the org’s connector tools as FunctionTools named postgres__query and so on; Deedglass runs each call. governTool puts Deedglass in front of a tool you wrote. There is no Python adapter yet: deedglass.openai_agents is a placeholder that raises ImportError.

examples/frameworks/openai-agents/native.ts
// The native route: @deedglass/sdk's OpenAI Agents adapter. Needs DEEDGLASS_API_KEY, DEEDGLASS_SERVER
// and OPENAI_API_KEY.
import { Agent, run } from '@openai/agents';
import { Deedglass } from '@deedglass/sdk';
import { deedglassTools } from '@deedglass/sdk/openai-agents';


// The org's connector tools, named postgres__query and so on; Deedglass runs each one.
const tools = await deedglassTools(new Deedglass(), { include: ['postgres.*'] });
const agent = new Agent({ name: 'orders', instructions: 'Answer questions about orders.', tools });
const result = await run(agent, 'How many rows are in public.orders?');
console.log(result.finalOutput);
Terminal window
npm install @openai/agents@0.21.0 zod@4.6.7 @deedglass/sdk@0.6.0
export DEEDGLASS_API_KEY="$(cat .deedglass/quickstart.key)" DEEDGLASS_SERVER=http://127.0.0.1:8080
node native.ts

Use the port init printed if it isn’t 8080.

Check it without a model key

Section titled “Check it without a model key”

examples/frameworks/openai-agents/ 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, lists its tools (Python) or connects (TypeScript), builds the Agent with the same mcp_config, and lists the native adapter’s tools. From the folder where init ran:

Terminal window
path/to/deedglass/examples/frameworks/run.sh openai-agents            # install the pins in a temp folder, run the check
path/to/deedglass/examples/frameworks/run.sh openai-agents --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:

  • A run with a live model (no model key in CI). The JavaScript docs don’t document a public listTools(), so the TypeScript check only connects.