Recipes: complete suites to start from
Each recipe is a whole project: a tamesu.yaml and a suite you can copy, change the URLs and
run. They validate as written (tamesu validate), and together they use most of what the
YAML reference lists. They are grouped by what is under test.
Chat and web agents
A REST agent gated for release, a web shop in the browser, the backend behind the transcript, and escalation.
- A support agent over REST, gated for release
- A web shop in the browser: mocks, accessibility, visual baseline
- Verify the backend, not just the transcript
- Escalation: when the agent must hand off, and when it must not
Voice agents
The OpenAI Realtime protocol, a phone IVR over SIP, a Voiceflow voice agent through its SIP connection, and Vapi, Deepgram and Retell agents on the generic targets.
- A voice agent on the OpenAI Realtime protocol
- A phone IVR through the SIP target
- A Vapi assistant over its websocket transport
- A Deepgram Voice Agent
- A Voiceflow voice agent, over SIP
- A Retell agent through a web call
Models, tools and agents
A prompt and its tools against a model, an MCP server through the model that uses it, and an orchestrator with the agents it delegates to.
- A prompt and its tools against a model, grounded and under failure
- An MCP server, through the model that uses it
- An orchestrator and the agents it delegates to
Messaging channels
Slack, Teams, Telegram and WhatsApp bots, and a Voiceflow agent, all over the REST target with asynchronous replies.
- A Slack bot, as a user in the workspace
- A Teams or Copilot Studio bot, through Direct Line
- A Telegram bot, with tamesu as the Bot API
- A WhatsApp bot, with tamesu as the Cloud API
- A Voiceflow agent over its Conversations API
Pipelines and operations
Red-team gating without a model key, a nightly red team, one suite for every environment and language, and a quarantined flaky test.