8616839cab
CHANGELOG promotion was a manual operator step that was never run, so 463 fragments for work already shipped in <=1.3.1 accumulated in .changeset/. Their notes were already hand-curated into the dated [1.2.0]/[1.3.0]/[1.3.1] CHANGELOG sections (#690 backfill, PR #694). Rendering them now would duplicate and mis-attribute shipped work. Move them to .changeset/archived/ (read non-recursively by all changeset tooling, so never rendered), keeping only the 3 genuinely-unreleased fragments at the top level. Prep for wiring `render` into the release finalize job (#690 follow-up). Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
603 B
603 B
type, pr
| type | pr |
|---|---|
| Fixed | 599 |
Large-file GSD writer agents no longer fail to write their output on OpenCode — gsd-research-synthesizer, gsd-planner, gsd-executor, gsd-domain-researcher, gsd-project-researcher, and gsd-ui-researcher now carry the same truncation-resilient write contract added for gsd-phase-researcher in #214. Each keeps its single-write default and falls back to an incremental, sentinel-based Write→Read→Edit sequence only when a runtime truncates an oversized tool call (upstream opencode#18108), instead of doom-looping on JSON parsing failed: Expected '}'.