opencode_env: run setup before the agent, skip verify on non-zero agent exit - #1023
opencode_env: run setup before the agent, skip verify on non-zero agent exit#1023sergiopaniego wants to merge 3 commits into
Conversation
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
![cursor[bot]](https://avatars.githubusercontent.com/in/1210556?s=60&v=4){kind=small}
run_rollout ran setup after create() had already started the agent, so setup raced the agent over the workspace. create() now takes start_agent (default True, so the loop-owning training path is unchanged), the server passes start_agent=False, runs setup, then calls session.start_agent(), giving the correct setup then agent then verify order. A non-zero agent exit is now treated like a timeout, so a crashed agent no longer runs verify or earns a reward. AI-assisted.
9975794 to
e7aadc8
Compare
There was a problem hiding this comment.
Cursor Bugbot has reviewed your changes using default effort and found 1 potential issue.
There are 2 total unresolved issues (including 1 from previous review).
❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.
Reviewed by Cursor Bugbot for commit e7aadc8. Configure here.
| _emit(f"setup FAILED at [{i}]: exit={cr.exit_code}") | ||
| break | ||
|
|
||
| # Block until the agent is done (or setup already failed). |
There was a problem hiding this comment.
Timeouts ignore sequential setup cost
Medium Severity
Deferring the agent until after setup makes wall-clock time setup + agent instead of overlapping them, but sandbox lifetime stays agent_timeout_s + 300 and the MCP tool cap stays 900s. With default agent_timeout_s=600 and SETUP_TIMEOUT_S=300, longer setups can kill the sandbox or trip the tool timeout before the agent finishes.
Additional Locations (2)
Reviewed by Cursor Bugbot for commit e7aadc8. Configure here.
Without it the run pipeline reports tee's exit code (0), so a non-zero opencode exit is masked and the failed-agent guard never fires. Matches the pi runtime, which already sets pipefail.
What does this PR do?
Two usability fixes to
opencode_env'srun_rollout, from review on #999.run_rolloutran the setup commands aftercreate(), which had already started the agent, so setup and the agent raced over the workspace.create()now takesstart_agent(defaultTrue, so the loop-owning training path that callscreate()directly is unchanged). The server passesstart_agent=False, runs setup, then callssession.start_agent(), giving the correct setup then agent then verify order.agent_exit_codeis now treated like a timeout (setsresult.error), so verify and reward are skipped for a crashed agent.The same fix for
pi_envis on #999.AI-assisted.
Note
Medium Risk
Changes rollout ordering and reward gating for the MCP server path; default
create()behavior is preserved but scoring semantics change when the agent crashes.Overview
run_rolloutnow runs setup before the OpenCode agent starts, instead of racing the agent over the workspace.OpenCodeSessionFactory.create()gains astart_agentflag (defaultTrueso direct training callers are unchanged); the MCP server usesstart_agent=False, runs per-command setup, then callssession.start_agent().Failed agent runs are no longer scored. A non-zero
agent_exit_codesetsresult.error(like timeouts), so verify and reward are skipped for a crashed agent.The agent launch shell command adds
set -o pipefailso failures in theopencode run | teepipeline surface as a non-zero exit. Factory lifecycle tests accept the newstart_agentparameter on mocked_create_once.Reviewed by Cursor Bugbot for commit 988f672. Bugbot is set up for automated code reviews on this repo. Configure here.