Mock environments for AI agent testing. Start with a fully seeded Gmail API — deterministic, resettable, and spec-compatible.
We are actively restructuring the repo to support more environments (Calendar, Drive, Slack) and adding reliability tests to existing ones.
pip install smolclawsSeed a Gmail environment with test data, then start the API server:
smolclaw seed --scenario default
smolclaw serve --port 8001 --no-mcpThe server exposes a Gmail-compatible REST API at http://localhost:8001/gmail/v1/.
Try it:
curl http://localhost:8001/gmail/v1/users/me/profile
curl http://localhost:8001/gmail/v1/users/me/messagesInteractive API docs are at http://localhost:8001/docs.
54 Gmail API endpoints — messages, threads, labels, drafts, settings, send-as, forwarding, delegates, vacation, filters, contacts, attachments.
Seedable scenarios — default (~57 emails across realistic threads), long_context (~3000 emails), or per-task scenarios.
State management — snapshot, diff, and restore. Every API call is logged for evaluation.
smolclaw seed --scenario default # seed + take initial snapshot
smolclaw reset # restore to initial stateAdmin API — inspect state, view action logs, compute diffs via /_admin/ endpoints.
| Scenario | Emails | Description |
|---|---|---|
default |
~57 | Standard inbox with threads, labels, attachments |
long_context |
~3000 | Stress test with high-volume realistic email |
smolclaw --db mydata.db seed # custom database path
smolclaw serve --host 0.0.0.0 # bind to all interfaces
smolclaw serve --port 9000 # custom portgit clone https://github.com/benchflow-ai/smolclaw.git
cd smolclaw
pip install -e ".[dev]"
pytest tests/MIT