Files
msd-core/.changeset
Tom Boucher 82fb847754 fix(#214): make gsd-phase-researcher survive OpenCode write-tool truncation (#598)
* chore: wire docs/agents config into AGENTS.md Agent skills section

Add the `## Agent skills` discovery block pointing the engineering
skills at the existing docs/agents/{issue-tracker,triage-labels,domain}.md
files (issue tracker, triage label mapping, single-context domain docs).

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

* fix(#214): make gsd-phase-researcher survive OpenCode write-tool truncation

OpenCode caps model output at OUTPUT_TOKEN_MAX=32000 and the thinking
budget shares that pool (upstream opencode#18108). A single oversized
`write` tool call for RESEARCH.md is truncated mid-payload, yielding
`JSON Parse error: Expected '}'`, which OpenCode misclassifies and then
doom-loops retrying identically. Short content writes fine; long
content fails 100% (reproducible, OpenCode 1.15.10).

Add a Step 6 write contract to agents/gsd-phase-researcher.md: keep the
single-Write default (no behavior change for Claude Code and other
runtimes that don't truncate), but on a truncation/invalid-tool failure
build the file incrementally via a sentinel-based Write -> Read -> Edit
sequence so no single tool-call payload is large enough to truncate;
never silently fall back to returning content (which truncates
identically). This is the upstream-recommended mitigation (write in
smaller chunks; use edit for follow-on writes).

Locked with a prompt-contract regression test mirroring the bug-222
write-contract pattern.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

* chore(#214): add changeset for OpenCode write-truncation fix

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
2026-06-01 21:42:08 -04:00
..

Changeset Fragments

This directory holds per-PR CHANGELOG fragments. Every PR with user-facing changes drops one (or more) <random-name>.md files here describing its CHANGELOG entry. Fragments are consolidated into the top-level CHANGELOG.md at release time.

Why

Two PRs that both edit the ### Fixed block of CHANGELOG.md always conflict on merge — git can't pick a serialization order without human input. Two PRs that each add a fresh .changeset/<unique-name>.md never conflict because they don't share lines.

See #2975 for the full rationale.

Adding a fragment

node scripts/changeset/new.cjs \
  --type Fixed \
  --pr 1234 \
  --body "fix the thing — explain the user-visible change in one sentence"

This writes .changeset/<adjective>-<noun>-<noun>.md with frontmatter and a body. Three random words → concurrent PRs don't collide.

Format

---
type: Fixed
pr: 1234
---
**`/gsd-foo` no longer drops trailing slashes** — explain the user-visible change.

Allowed type: values follow Keep a Changelog: Added, Changed, Deprecated, Removed, Fixed, Security.

Opting out

PRs that legitimately have no user-facing impact can add the no-changelog label. CI honors it. When unsure, add the fragment.

At release time

node scripts/changeset/cli.cjs render --version vX.Y.Z --date YYYY-MM-DD

Reads every fragment, groups bullets by type:, replaces ## [Unreleased] with a new ## [vX.Y.Z] - YYYY-MM-DD block, opens a fresh ## [Unreleased] above, deletes consumed fragments. Idempotent.