When working on a feature branch (e.g. f-123-new-thing), GSD creates
slice branches correctly from the current branch but merges them back
to main instead of the feature branch. This is because getMainBranch()
always resolved to the repo default branch with no concept of where
the user started.
Fix: record the current branch as the "integration branch" in a
per-milestone metadata file (.gsd/milestones/<MID>/<MID>-META.json)
when auto-mode starts. getMainBranch() checks this metadata before
falling back to repo defaults, so switchToMain() and mergeSliceToMain()
target the correct branch.
Key details:
- Integration branch is captured once per milestone (idempotent)
- Committed immediately so it survives branch switches (.gsd/ files
are discarded during checkout)
- main_branch preference still takes highest priority
- Falls back to existing detection if metadata missing (backward compat)
- Per-milestone: different milestones can target different branches
- Validates branch still exists before using it
Tests: 41 new assertions across git-service.test.ts and worktree.test.ts
covering the full lifecycle, multi-slice workflows, resume scenarios,
backward compatibility, and edge cases.
In RPC mode, `ctx.ui.custom()` returns `undefined as never`, causing
`showInterviewRound` to return undefined and `Object.keys(result.answers)`
to throw TypeError.
When `showInterviewRound` returns undefined (RPC mode), fall back to
sequential `ctx.ui.select()` calls for each question, forwarding the
abort signal (#171) and supporting `allowMultiple` (#165).
- Add `allowMultiple` to `ExtensionUIDialogOptions`
- Widen `select()` return type to `string | string[] | undefined`
- Add `allowMultiple` to RPC select request and `values` array to response
- Update RPC `select()` to forward `allowMultiple` and parse array responses
- Guard existing `ctx.ui.select()` callers against the widened return type
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Auto mode's model resolution used `allModels.find(m => m.id === modelId)`
which returns the first match regardless of provider. With 30+ duplicate
model IDs across providers, user preferences silently resolved to the
wrong provider.
Three fixes:
- Use `getAvailable()` instead of `getAll()` so only authenticated
models are considered
- Support `provider/model` format (e.g. "google/gemini-2.5-pro") for
explicit provider targeting
- For bare IDs, prefer the current session's provider, then first
available match, with an ambiguity warning
- Store and restore original model provider instead of hardcoding
"anthropic" when restoring the user's model after auto-mode
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
verifyExpectedArtifact only checked for the task summary file, but
deriveState determines the next task by finding the first unchecked
checkbox in the slice plan. When the agent writes the summary but
doesn't mark the checkbox, the dispatch loop re-sends the same unit
and gets stuck after max retries.
Part 1: verifyExpectedArtifact now also checks that the task checkbox
is marked [x] in the slice plan for execute-task units.
Part 2: At retry time, if the summary exists but the checkbox is
unmarked, the dispatch logic self-repairs by marking the checkbox
programmatically (via skipExecuteTask) and re-derives state instead
of re-dispatching the same unit.
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Bug 1: resolveMilestoneFile/resolveSliceFile already check file existence
via readdirSync, so the additional loadFile content check was redundant.
Empty research files (exists on disk but no content) caused a loop where
verifyExpectedArtifact marked research complete while dispatch re-triggered
it because loadFile returned falsy for empty files.
Bug 2: stopAuto now calls rebuildState to synchronize disk state, matching
pauseAuto's pattern and preventing stale state on next resume.
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Auto mode exits when mergeSliceToMain() hits conflicts on runtime files
like completed-units.json that were manually committed via `gsd queue`.
Two-part fix:
- Untrack RUNTIME_EXCLUSION_PATHS from the index before merge starts
- If merge conflicts are limited to runtime files, auto-resolve by
taking ours and removing from index instead of aborting
Closes#189
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
The before_provider_request hook used model.startsWith("claude") to gate
native web search injection. This matched claude-* models served by any
provider (GitHub Copilot, AWS Bedrock, etc.), incorrectly injecting
Anthropic-only web_search_20250305 tool definitions into non-Anthropic
API requests.
The fix checks the isAnthropicProvider flag (set by model_select via the
provider field) instead of sniffing the model name.
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Three locations used lastIndexOf("/") or includes("/") for path
manipulation, which fails on Windows where paths use backslashes.
- auto.ts: writeBlockerPlaceholder directory extraction → dirname()
- interactive-mode.ts: parent directory traversal → path.dirname() loop
- path-utils.ts: non-null assertion on MSYS drive letter access
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
On Windows, LLMs convert absolute paths like F:\Projects\.gsd\... to
Unix-style /f/Projects/.gsd/... which Node's path.resolve interprets
as drive-root-relative, creating F:\f\Projects\.gsd\... instead.
Replace all *AbsPath template variables in prompt templates with
relative .gsd/... paths that resolve correctly on all platforms.
Add MSYS path normalization in resolveToCwd as defense-in-depth.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Repos that started tracking .gsd/completed-units.json before the
gitignore rule was added continue to see squash-merge conflicts because
.gitignore only prevents new tracking. This adds a bootstrap step that
runs `git rm --cached` on all RUNTIME_EXCLUSION_PATHS, eliminating the
conflict at its source.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
LLM-generated commands with `> NUL` create undeletable files on Windows
because Git Bash treats NUL as a literal filename. Rewrite NUL redirects
to /dev/null at all three bash spawn sites.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
emitObservabilityWarnings only called ctx.ui.notify — the agent never
saw the warnings and ignored them entirely. Validator caught real issues
(missing observability sections, placeholder diagnostics) but had zero
enforcement.
Rename to collectObservabilityWarnings (returns issues), add
buildObservabilityRepairBlock to format issues as actionable prompt
instructions. Appended to the unit prompt so the agent reads flagged
files and fixes gaps before proceeding with the unit.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Browser launch was hardcoded to headless: false, crashing on Linux
servers without a display server ($DISPLAY). Auto-detect headless
environments and also support FORCE_HEADLESS=true override.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Squash merge was hardcoded, causing auto-mode to hard-stop when conflicts
arose from long-lived branches or frequently-changing .gsd/* artifacts.
Add git.merge_strategy preference ("squash" | "merge", default: squash).
"merge" uses --no-ff which preserves branch history and handles conflicts
from divergent branches more gracefully. Users hitting repeated squash
merge failures can set merge_strategy: merge in .gsd/preferences.md.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
complete-slice verification only checked for the SUMMARY file, so when
the LLM skipped writing the UAT, the unit was marked complete and UAT
was never produced. Users saw doctor-created placeholder UATs instead
of real test scripts.
- verifyExpectedArtifact now checks both SUMMARY and UAT for complete-slice
- complete-slice prompt strengthened: step 7 requires concrete test cases,
MUST line lists all three required artifacts with enforcement warning
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Three-layer fix for runtime files leaking into git commits:
1. Stage-then-unstage: replace pathspec excludes with git add -A followed
by git reset HEAD for each exclusion. The old approach failed when .gsd/
was in .gitignore — git exited non-zero before evaluating the excludes,
and the catch fallback staged everything unconditionally.
2. Auto-cleanup: on first smartStage call per session, remove any runtime
files that are already tracked in the index (from the fallback bug) via
a dedicated commit. This is a one-time migration that self-heals repos
where runtime files were accidentally committed.
3. Pre-checkout discard: after pre-switch auto-commits that exclude .gsd/,
run git checkout -- .gsd/ to clear dirty runtime files that would
otherwise block git checkout when the target branch has different
tracked versions.
Also adds completed-units.json to RUNTIME_EXCLUSION_PATHS and
BASELINE_PATTERNS (was missing — metrics.json was listed but
completed-units.json was not).
Tasks:
- chore(M002/S06): auto-commit after complete-slice
- chore(M002/S06): auto-commit after complete-slice
- chore(M002/S06/T02): auto-commit after execute-task
- chore(M002/S06/T02): auto-commit after execute-task
- chore(M002/S06/T01): auto-commit after execute-task
- chore(M002/S06/T01): auto-commit after execute-task
- chore(M002/S06): auto-commit after plan-slice
- chore: update state for S06 execution
- docs(S06): add slice plan
Branch: gsd/M002/S06
Tasks:
- chore(M002/S02): auto-commit after complete-slice
- chore(M002/S02): auto-commit after complete-slice
- chore(M002/S02/T02): auto-commit after execute-task
- chore(M002/S02/T02): auto-commit after execute-task
- chore(M002/S02/T01): auto-commit after execute-task
- fix: expand tool outputs by default and pull main before slice merge
- chore(M002/S02): auto-commit after plan-slice
- docs(S02): add slice plan
Branch: gsd/M002/S02
interview-ui.ts saveEditorToState() was calling getText() which returns
paste markers like '[paste #1 2033 chars]' for content >1000 chars or
>10 lines. The actual pasted content was stored in the Editor's paste
map but never expanded back.
This silently discards user input in ask_user_questions notes — any
substantive response (voice transcripts, detailed explanations, extended
enrichments) that exceeds the paste threshold gets replaced with a
marker string. The LLM receives '[paste #1 N chars]' instead of the
user's actual words.
Fix: getText() → getExpandedText() — the Editor already has the method
that expands paste markers to their stored content. One-line change.
- /exit now calls stopAuto() before exiting to save activity log and clear locks
- Added new /kill command for immediate exit without cleanup
- Fixes issue #132: /exit terminates too abruptly and leaves terminal state dirty
The ESM resolve hook was rewriting .js imports from vendored Pi
packages (packages/*/dist/) to .ts, breaking test resolution.
Compiled dist/ files need their .js specifiers left intact.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Vendor all 4 Pi packages (tui, ai, agent-core, coding-agent) from
pi-mono v0.57.1 as @gsd/* workspace packages under packages/. This
replaces the compiled npm dependency (@mariozechner/pi-coding-agent)
and patch-package workflow, giving direct source access for
modifications.
- Copy Pi source from pi-mono v0.57.1 into packages/
- Create workspace package.json + tsconfig.json for each package
- Rename ~240 imports from @mariozechner/pi-* to @gsd/pi-*
- Apply existing patches as source edits (setModel persist, VT input)
- Remove @mariozechner/pi-coding-agent dep and patch-package
- Update build pipeline to build packages in dependency order
- Add pi-upstream git remote for future selective syncing
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Doctor's fix:true mode was creating summary stubs and marking slices
done in the roadmap during the post-hook after every task. This
short-circuited the complete-slice dispatch unit — by the time
dispatchNextUnit ran, the slice was already 'done' and the merge
guard merged it to main, so complete-slice (which writes the real
compressed summary) never got a chance to run.
Root cause: doctor conflated two responsibilities — task-level
bookkeeping (marking checkboxes) and completion state transitions
(summary stubs, roadmap marking). The post-hook should only do
the former.
Fix: added fixLevel option to runGSDDoctor. fixLevel:'task' (used
by post-hook) skips completion transition codes. fixLevel:'all'
(default, used by manual gsd doctor and resume) preserves existing
recovery behavior.
Completion transition codes gated by fixLevel:
- all_tasks_done_missing_slice_summary
- all_tasks_done_missing_slice_uat
- all_tasks_done_roadmap_not_checked
Restores main_branch field on GitPreferences (removed in a prior merge conflict
resolution) and adds VALID_BRANCH_NAME validation in getMainBranch(). Implements
runPreMergeCheck with auto-detection from package.json test scripts and support
for custom commands via prefs.pre_merge_check string values.
Fixes 5 pre-existing test failures in git-service.test.ts (158/158 now pass).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The crash loop: stale state → unit redispatched → activity log grows →
retry diagnostic reads full log → prompt grows → replaceAll on huge
string → V8 heap exhaustion. Cap both the read path (10MB JSONL parse
limit) and the injection path (50K char prompt cap) to break the cycle.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Pre-switch auto-commits were including .gsd/ planning artifacts (roadmaps, STATE.md)
on both sides of a branch switch, causing reliable merge conflicts when squash-merging
slices back to main. Now pre-switch auto-commits exclude the entire .gsd/ directory,
while post-task auto-commits continue to include them normally.
Also restores VALID_BRANCH_NAME export removed in a prior merge conflict resolution.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Front-load API key collection into GSD's planning phase so auto-mode
runs uninterrupted. Planning prompts forecast secrets into a manifest,
auto-mode collects pending keys before dispatching the first slice.
- getManifestStatus() queries manifest state against env
- collectSecretsFromManifest() orchestrates summary, collection, manifest update
- showSecretsSummary() read-only TUI summary with status indicators
- collectOneSecret() enhanced with guidance display above masked input
- Secrets gate in startAuto() — non-fatal, inherited by guided flow
- 19 new tests (manifest-status, collect-from-manifest, auto-secrets-gate)
- All 10 requirements (R001-R010) validated
mergeSliceToMain now runs git reset --hard if git merge --squash fails,
restoring a clean working tree instead of leaving conflict markers.
The merge guard catch block in auto.ts now:
1. Detects leftover conflicted state (UU/AA/UD in porcelain status)
2. Resets the working tree if conflicts remain
3. Stops auto-mode with a clear error instead of continuing with
corrupted .gsd/ state files that cause an infinite dispatch loop
Also fixes conflict markers in loader.ts, logo.ts, and postinstall.js
that were baked into main from a prior bad merge resolution.
