Skip to content

Commit d81bdef

Browse files
mp-orkesclaude
andcommitted
Fix stale docstring in example 82_fan_out_fan_in
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
1 parent c10590e commit d81bdef

1 file changed

Lines changed: 6 additions & 3 deletions

File tree

examples/agents/82_fan_out_fan_in.py

Lines changed: 6 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -21,16 +21,19 @@
2121
- Filesystem IPC:
2222
* Workers write sentinels after submit_answer so main counts completions
2323
* Collector writes reports to files; main thread reads and prints them
24-
- No time.sleep() to assume message delivery — synchronisation via files
24+
- Result delivery is never assumed from elapsed time — every wait is on a
25+
sentinel file. The one bare sleep is a startup grace period, letting the
26+
agents reach their first wait call before the first message is sent.
2527
2628
Scenario:
2729
A research Orchestrator fans out each question to three Worker agents
2830
(alpha, beta, gamma) that produce independent short answers. The Collector
2931
aggregates the three answers into a side-by-side comparison report.
3032
3133
Requirements:
32-
- Conductor server running (CONDUCTOR_SERVER_URL / CONDUCTOR_SERVER_URL)
33-
- CONDUCTOR_AGENT_LLM_MODEL set to a working model
34+
- Conductor server with WMQ support (conductor.workflow-message-queue.enabled=true)
35+
- CONDUCTOR_SERVER_URL=http://localhost:8080/api as environment variable
36+
- CONDUCTOR_AGENT_LLM_MODEL=openai/gpt-4o-mini as environment variable
3437
"""
3538

3639
import json

0 commit comments

Comments
 (0)