fix(test): raise e2e_01 task_timeout to 2400s - #1981
dushyant-uipath wants to merge 1 commit into
Conversation
The nightly default is 1200s total. e2e_01 prompts for SharePoint and SAP which triggers real connector registry lookups — these eat the clock before the HITL node is ever added. 2400s matches the existing per-turn limit and gives the agent enough headroom. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
Claude finished @dushyant-uipath's task in 48s —— View job PR Review: fix(test): raise e2e_01 task_timeout to 2400s
SummaryAdds the missing Change-by-Change Review1.
|
| Area | Status | Notes |
|---|---|---|
| Frontmatter | N/A | No skill frontmatter changed |
| E2E Tests | OK | Existing test task updated with correct timeout |
| Skill Body | N/A | No skill body changed |
| References & Assets | N/A | No references or assets changed |
| Repo Hygiene | OK | Scoped single-line fix, no secrets or stray changes |
Issues for Manual Review
None found.
Conclusion
Clean, minimal, well-motivated fix. The root cause is clearly explained in the PR description and the change directly addresses it. Approve.
|
Claude finished @dushyant-uipath's task in 1m 47s —— View job Coder-eval task lint (advisory)1 task YAML changed; verdict counts: 0 Critical, 0 High, 1 Medium, 0 Low, 0 OK. Rubric: .claude/commands/lint-task.md. This check is advisory and never blocks merge. Evidence of passing run❌ High — PR body does not claim the modified task has been re-run and still passes with the new
Per-task lint
|
e2e_01mentions SharePoint and SAP in the prompt. The agent does real connector registry lookups for both, which burns most of the 1200s nightly default before the HITL node is ever placed. The flow produced has no HITL node and no edges — score 0.0, timeout.Root cause:
turn_timeout: 2400was already set (per-turn) buttask_timeoutwas missing, so the 1200s nightly default applied to the whole task.Fix: add
task_timeout: 2400to match the per-turn limit.🤖 Generated with Claude Code