Cursor Agent SDK vs guilds.run

Cursor Agent SDK

An agent-as-code framework with typed tools, declared read and write effects that drive retries and dry runs, an approval gate per tool and evals that assert on tool calls. It's closed source, every model turn runs through Cursor's backend on a Cursor plan, local sandboxing is off by default, and the changelog has two entries.

guilds.run

A self-hosted multi-agent system that records each run's prompt, tool calls and memory changes, with per-agent models and tools and no telemetry by default. Public since 9 October 2026, it has no tagged release or published package, its own security register lists 13 open issues, and the site's terms and privacy links return 404.

Cursor Agent SDK

1. Install @grok/agent-sdk, not @cursor/agent-sdk from the launch post. That name answers 404 on npm
2. Set local.sandbox before a turn reads webhook payloads, pull request diffs or chat. Without it shell commands auto-approve with the host's credentials
3. Give every tool an effect. A tool without one counts as a write, runs once and is stubbed in dry runs
4. Start agent-sdk serve with --bearer-token on any shared host. --allow-anonymous admits every caller as one principal
5. Delete bot/hooks/memory.ts when sessions serve different people. The journal shares each turn's message and result with later turns

guilds.run

1. Keep the box on loopback or behind an authenticating proxy with AUTH_MODE=trusted-header. The default mode does not authenticate
2. Put agents that hold different secrets or GitHub tokens on different machines. Agents on one machine can read each other's homes and tokens
3. Back up the .env file written by ./setup.sh. It holds the only copy of the key that encrypts stored credentials
4. Set a spend limit at your model provider. guilds.run sets no spend cap, and max_tool_calls defaults to 50 per run
5. Pin a commit of main when you deploy. There are no tagged releases or versioned images yet