Evidence record / omp
A dated record of one change, kept so the writing that cites it can be checked. Compiled from the sources listed below by the research run, not written for reading. The judgment lives in the issues that cite it, below.
2026-10-13-omp-task-subagents-switched-to-plain-text-results-by-default-while-eval-spawns-stayed-structur
Task subagents switched to plain-text results by default, while eval spawns stayed structured. task.structured defaults false: the task tool hides outputSchema/schemaMode, removes schema and yield injection from its children, and returns the child’s final assistant message as plain text. Enabling the setting restores validated task-result schemas. Eval agent() and workpool() spawns keep structured results independently.
Channel: tagged-release. Half: both. Date: 2026-10-12 (v18.10.2).
Operator consequence: Enable task.structured explicitly for task consumers requiring schema-validated output; do not assume the same default affects eval spawns. Check consumers before upgrading a fleet.
Receipt
Finding metadata
Run: 2026-10-13-profile-refresh-2026-09-21_2026-10-13-frontier-v0
Finding ID: 2026-10-13-omp-task-subagents-switched-to-plain-text-results-by-default-while-eval-spawns-stayed-structur
Profile citations
- omp / claim / plain-task-results
Source links
Primary links, including exact changelog lines when available.
- commit2026-10-13-omp-task-subagents-switched-to-plain-text-results-by-default-while-eval-spawns-stayed-structurgithub.com/can1357/oh-my-pi/commit/3ad2b788d7a078ec79a6962f5c9e690d00bbb268tagged commit file2026-10-13-omp-task-subagents-switched-to-plain-text-results-by-default-while-eval-spawns-stayed-structurcan1357/oh-my-pi / packages/coding-agent/src/task/settings.ts#L255-L265github release2026-10-13-omp-task-subagents-switched-to-plain-text-results-by-default-while-eval-spawns-stayed-structurcan1357/oh-my-pi / v18.10.2
Versioned source: run artifact