Two agents connected by an edge. The first classifies the user’s intent; the second reads that classification from the workflow context and composes a reply. This classify → respond pattern appears in nearly every customer-facing agent workflow.
What it demonstrates
- Connecting two nodes with an
edge
- Writing intermediate results to
working.* context paths
- Mixing providers: OpenAI classifies, Anthropic replies
- Workflow-level
guardrails
Run it
sirenspec run docs/cookbook/sequential-pipeline/workflow.yaml --input "My order hasn't arrived yet."
Workflow
docs/cookbook/sequential-pipeline/workflow.yaml
version: "0.1"
agents:
classifier:
model: "openai:gpt-4o-mini"
system: |
Classify the user's message into one of these intents: question, complaint, feedback, other.
Respond with ONLY the intent label, nothing else.
replier:
model: "anthropic:claude-haiku-4-5-20251001"
system: |
You are a customer support agent. The user's intent has been classified.
Compose a short, helpful reply appropriate for that intent.
nodes:
classify:
agent: classifier
writes: working.intent
reply:
agent: replier
writes: output.reply
edges:
- from: classify
to: reply
guardrails:
- injection
- length
This workflow mixes providers: OpenAI handles classification, Anthropic writes the reply. Each agent independently resolves its own provider at runtime.
How data flows
classify receives the user’s message and writes a single intent label (e.g. "complaint") to working.intent.reply receives the output of classify as its user message and composes the final response.
Graph
Next steps
Conditional Pipeline
Route to different handlers based on the classification.
Telephone Game
Chain five agents and watch meaning drift.
