gsdd-cli 0.27.0 → 0.29.0

package/distilled/workflows/progress.md

@@ -7,25 +7,25 @@ Scope boundary: you are NOT resume.md. You do not wait for user input, clean up
 </role>
 
 <control_map>
-At the start of status reporting, run `node .planning/bin/gsdd.mjs control-map --json` when the local helper exists. Summarize its computed repo/worktree/planning state in the status block: canonical branch/HEAD, tracked/untracked dirty buckets, whether ignored paths were scanned, sibling or detached worktrees, stale local annotations, planning drift, and recommended interventions. Use `--with-ignored` before making a clean-workspace claim that includes ignored or generated surfaces. Treat the command output as read-only computed evidence. Local annotations under `.planning/.local/` explain intent but never outrank repo truth, planning artifacts, or checkpoint reconciliation.
+At the start of status reporting, run `node .work/bin/gsdd.mjs control-map --json` when the local helper exists. Summarize its computed repo/worktree/planning state in the status block: canonical branch/HEAD, tracked/untracked dirty buckets, whether ignored paths were scanned, sibling or detached worktrees, stale local annotations, planning drift, and recommended interventions. Use `--with-ignored` before making a clean-workspace claim that includes ignored or generated surfaces. Treat the command output as read-only computed evidence. Local annotations under `.work/.local/` explain intent but never outrank repo truth, planning artifacts, or checkpoint reconciliation.
 </control_map>
 
 <prerequisites>
-`.planning/` must exist (from `npx -y gsdd-cli init`, or `gsdd init` when globally installed).
+`.work/` must exist (from `npx -y gsdd-cli init`, or `gsdd init` when globally installed).
 
-This is a read-only workflow. No files are created, modified, or deleted. If `.planning/` does not exist, tell the user to run `npx -y gsdd-cli init` and stop.
+This is a read-only workflow. No files are created, modified, or deleted. If `.work/` does not exist, tell the user to run `npx -y gsdd-cli init` and stop.
 </prerequisites>
 
 <repo_root_helper_contract>
-All `node .planning/bin/gsdd.mjs ...` helper references below assume the current working directory is the repo root. If the runtime launched from a subdirectory, change to the repo root before acting on them.
+All `node .work/bin/gsdd.mjs ...` helper references below assume the current working directory is the repo root. If the runtime launched from a subdirectory, change to the repo root before acting on them.
 </repo_root_helper_contract>
 
 <lifecycle_boundary>
 `progress` stays read-only.
 
 - Derive lifecycle posture from repo truth only; do not mutate phase or milestone state from this workflow.
-- Do not call `node .planning/bin/gsdd.mjs phase-status` here.
-- If you recommend a next step that crosses a lifecycle boundary, the downstream mutating workflow must rerun its own `node .planning/bin/gsdd.mjs lifecycle-preflight ...` gate before acting.
+- Do not call `node .work/bin/gsdd.mjs phase-status` here.
+- If you recommend a next step that crosses a lifecycle boundary, the downstream mutating workflow must rerun its own `node .work/bin/gsdd.mjs lifecycle-preflight ...` gate before acting.
 </lifecycle_boundary>
 
 <process>
@@ -33,12 +33,13 @@ All `node .planning/bin/gsdd.mjs ...` helper references below assume the current
 <check_existence>
 Check for project artifacts in order:
 
-1. **No `.planning/` directory** — tell the user to run `npx -y gsdd-cli init`. Stop.
-2. **If `.planning/brownfield-change/CHANGE.md` exists** — treat this as the active medium-scope brownfield continuity state. Go to Branch F.
-3. **No `.planning/ROADMAP.md` AND no `.planning/SPEC.md`** — check for non-phase brownfield artifacts:
-   - if `.planning/codebase/` has substantive map documents, or `.planning/quick/` has LOG/task artifacts, treat this as a non-phase brownfield state. Go to Branch F.
+1. **No `.work/` directory** — tell the user to run `npx -y gsdd-cli init`. Stop.
+2. **If `.work/brownfield-change/CHANGE.md` exists and `Current posture` is not `closed`** — treat this as the active medium-scope brownfield continuity state. Go to Branch F.
+   - If `Current posture` is `closed`, keep the file as historical context only and continue checking ROADMAP/SPEC state.
+3. **No `.work/ROADMAP.md` AND no `.work/SPEC.md`** — check for non-phase brownfield artifacts:
+   - if `.work/codebase/` has substantive map documents, or `.work/quick/` has LOG/task artifacts, treat this as a non-phase brownfield state. Go to Branch F.
    - otherwise the project has no artifacts. Suggest running the `/gsdd-new-project` workflow. Stop.
-4. **No `.planning/ROADMAP.md` BUT `.planning/SPEC.md` exists** — this is a between-milestones state (milestone was completed and archived). Go to Branch F.
+4. **No `.work/ROADMAP.md` BUT `.work/SPEC.md` exists** — this is a between-milestones state (milestone was completed and archived). Go to Branch F.
 5. **Both exist** — proceed to derive status, including whether a retained `ROADMAP.md` already represents an archived milestone rather than an audit-ready one.
 </check_existence>
 
@@ -46,27 +47,27 @@ Check for project artifacts in order:
 Read the following and extract state:
 
 **Project identity:**
-- If `.planning/SPEC.md` exists, read it and extract the project name from the first heading.
-- If `.planning/SPEC.md` does not exist, derive the project name from the repo root directory name.
+- If `.work/SPEC.md` exists, read it and extract the project name from the first heading.
+- If `.work/SPEC.md` does not exist, derive the project name from the repo root directory name.
 
 **Non-phase brownfield state:**
-If `.planning/ROADMAP.md` does not exist, determine whether the repo is currently in one of these Branch F states:
-- `active_brownfield_change` — `.planning/brownfield-change/CHANGE.md` exists; read `CHANGE.md` first as the canonical operational anchor, then read `HANDOFF.md` for judgment-only context
-- `between_milestones` — `.planning/SPEC.md` exists
-- `codebase_only` — `.planning/codebase/` has substantive map documents but `.planning/SPEC.md` does not exist
-- `quick_lane` — `.planning/quick/LOG.md` or quick task directories exist but `.planning/SPEC.md` and `.planning/ROADMAP.md` do not
+If `.work/ROADMAP.md` does not exist, determine whether the repo is currently in one of these Branch F states:
+- `active_brownfield_change` — `.work/brownfield-change/CHANGE.md` exists and `Current posture` is not `closed`; read `CHANGE.md` first as the canonical operational anchor, then read `HANDOFF.md` for judgment-only context
+- `between_milestones` — `.work/SPEC.md` exists
+- `codebase_only` — `.work/codebase/` has substantive map documents but `.work/SPEC.md` does not exist
+- `quick_lane` — `.work/quick/LOG.md` or quick task directories exist but `.work/SPEC.md` and `.work/ROADMAP.md` do not
 
 For `active_brownfield_change`, `codebase_only`, and `quick_lane`, there is no active phase count. Record the non-phase state instead of trying to infer current milestone progress.
 
 **Active brownfield change:**
-If `.planning/brownfield-change/CHANGE.md` exists, extract:
+If `.work/brownfield-change/CHANGE.md` exists and is not closed, extract:
 - change title from the first heading
 - current posture from `## Current Status`
 - current branch / integration surface from `## Current Status`
 - next action from `## Next Action`
 - declared write scope from `## PR Slice Ownership` when present
 
-If `.planning/brownfield-change/HANDOFF.md` exists, read it as judgment-only context:
+If `.work/brownfield-change/HANDOFF.md` exists, read it as judgment-only context:
 - active constraints
 - unresolved uncertainty
 - decision posture
@@ -75,7 +76,7 @@ If `.planning/brownfield-change/HANDOFF.md` exists, read it as judgment-only con
 Do not treat `HANDOFF.md` as a co-equal status source. It explains the active change; `CHANGE.md` remains the operational anchor.
 
 **Phase statuses:**
-If `.planning/ROADMAP.md` exists, read it and parse phase statuses:
+If `.work/ROADMAP.md` exists, read it and parse phase statuses:
 - `[ ]` = not started
 - `[-]` = in progress
 - `[x]` = done
@@ -89,29 +90,29 @@ Determine:
 **Archived milestone evidence:**
 If `ROADMAP.md` exists and all phases in the current milestone are `[x]`, determine whether this is still audit-ready or already archived-with-`ROADMAP.md`:
 - derive the current milestone/version from the active milestone heading in `ROADMAP.md`
-- check `.planning/MILESTONES.md` for a shipped entry matching that same milestone/version
-- check for the matching archived milestone audit artifact for that same milestone/version (for example `.planning/v1.1-MILESTONE-AUDIT.md`)
+- check `.work/MILESTONES.md` for a shipped entry matching that same milestone/version
+- check for the matching archived milestone audit artifact for that same milestone/version (for example `.work/v1.1-MILESTONE-AUDIT.md`)
 - if both the shipped ledger entry and the matching archived audit artifact exist, treat the retained `ROADMAP.md` as archived milestone evidence and route to Branch F instead of Branch E
 - if either one is missing, keep the milestone in the audit-ready Branch E state
 
 **Checkpoint:**
-Check if `.planning/.continue-here.md` exists. If yes, note the `workflow` and `phase` frontmatter and the `next_action` section.
+Check if `.work/.continue-here.md` exists. If yes, note the `workflow` and `phase` frontmatter and the `next_action` section.
 - Treat checkpoint routing classes explicitly:
   - `phase` and `quick` checkpoints remain blocking resume-owned surfaces for routing only when there is no active brownfield change, or when a shared strict-match rule proves they still describe the active execution surface.
   - `generic` checkpoints are informational-only for this read-only reporter: show the checkpoint and its `next_action`, but keep evaluating the real lifecycle recommendation instead of routing Branch A back through `/gsdd-resume`.
-- If `.planning/brownfield-change/CHANGE.md` also exists, apply one shared strict-match rule before letting a surviving `phase` or `quick` checkpoint outrank the operational anchor:
+- If `.work/brownfield-change/CHANGE.md` also exists, apply one shared strict-match rule before letting a surviving `phase` or `quick` checkpoint outrank the operational anchor:
   - branch alignment: the checkpoint branch, `CHANGE.md` integration surface, and current git branch all match
   - scope alignment: the live dirty tree stays inside the declared brownfield write scope
   - still-active execution state: the checkpoint still points at live unfinished `phase` or `quick` work
 - If any one of those checks fails, keep the checkpoint visible in the status block but continue routing from the active brownfield change instead of bouncing Branch A back through `/gsdd-resume`.
 
 **Incomplete work:**
-If `.planning/phases/` exists, scan it for:
+If `.work/phases/` exists, scan it for:
 - PLAN files without a matching SUMMARY file (incomplete execution)
-- SUMMARY files without a matching VERIFICATION file (unverified, only relevant if `workflow.verifier` is enabled in `.planning/config.json`; if config.json cannot be read, assume verifier is disabled)
+- SUMMARY files without a matching VERIFICATION file (unverified, only relevant if `workflow.verifier` is enabled in `.work/config.json`; if config.json cannot be read, assume verifier is disabled)
 
 **Quick task log:**
-If `.planning/quick/LOG.md` exists, check the last entry for a non-terminal status.
+If `.work/quick/LOG.md` exists, check the last entry for a non-terminal status.
 
 **Artifact-versus-worktree mismatch:**
 If an active brownfield change exists, compare `CHANGE.md` to live git/worktree truth:
@@ -129,7 +130,7 @@ Run `git log main..HEAD --oneline` to detect commits on the current branch that
 </derive_status>
 
 <recent_work>
-Scan `.planning/phases/` for the 2-3 most recent SUMMARY.md files (by directory name or file modification time).
+Scan `.work/phases/` for the 2-3 most recent SUMMARY.md files (by directory name or file modification time).
 
 For each, extract:
 - Phase name from the directory name (e.g., `01-setup` → "Phase 1: Setup")
@@ -256,7 +257,7 @@ Suggested next action:
 ```
 
 **Branch E: Audit milestone (all phases [x], not yet archived)**
-Condition: All phases in the current milestone are marked `[x]`, and the current roadmap milestone/version does **not** yet have both a shipped entry in `.planning/MILESTONES.md` and the matching archived milestone audit artifact.
+Condition: All phases in the current milestone are marked `[x]`, and the current roadmap milestone/version does **not** yet have both a shipped entry in `.work/MILESTONES.md` and the matching archived milestone audit artifact.
 
 ```
 Suggested next action:
@@ -266,26 +267,26 @@ Suggested next action:
 
 **Branch F: Non-phase state (no active roadmap, or retained roadmap already archived)**
 Condition:
-- `.planning/brownfield-change/CHANGE.md` exists, **or**
-- `.planning/SPEC.md` exists but `.planning/ROADMAP.md` does not, **or**
-- `.planning/codebase/` or `.planning/quick/` exists while both `.planning/SPEC.md` and `.planning/ROADMAP.md` are absent, **or**
-- `.planning/ROADMAP.md` still exists, but the current roadmap milestone/version already has both a shipped entry in `.planning/MILESTONES.md` and the matching archived milestone audit artifact — this is the archived-with-`ROADMAP.md` state, not a second trip through audit
+- `.work/brownfield-change/CHANGE.md` exists, **or**
+- `.work/SPEC.md` exists but `.work/ROADMAP.md` does not, **or**
+- `.work/codebase/` or `.work/quick/` exists while both `.work/SPEC.md` and `.work/ROADMAP.md` are absent, **or**
+- `.work/ROADMAP.md` still exists, but the current roadmap milestone/version already has both a shipped entry in `.work/MILESTONES.md` and the matching archived milestone audit artifact — this is the archived-with-`ROADMAP.md` state, not a second trip through audit
 
-Check `.planning/MILESTONES.md`:
+Check `.work/MILESTONES.md`:
 - If MILESTONES.md exists and has at least one milestone entry → this is a subsequent milestone
 - If MILESTONES.md does not exist or is empty → this is the first milestone setup
 
 ```
 Suggested next action (active brownfield change):
-  Run /gsdd-resume to restore the active brownfield change context from `.planning/brownfield-change/CHANGE.md`
-  Also available: inspect `.planning/brownfield-change/HANDOFF.md`, /gsdd-progress (refresh after the artifact or worktree changes), /gsdd-new-project (only if you intentionally want to widen this bounded change into the first milestone), /gsdd-new-milestone (only if the repo already has shipped milestone history and you intentionally want to widen this change into the next milestone cycle)
+  Run /gsdd-resume to restore the active brownfield change context from `.work/brownfield-change/CHANGE.md`
+  Also available: inspect `.work/brownfield-change/HANDOFF.md`, /gsdd-progress (refresh after the artifact or worktree changes), /gsdd-new-project (only if you intentionally want to widen this bounded change into the first milestone), /gsdd-new-milestone (only if the repo already has shipped milestone history and you intentionally want to widen this change into the next milestone cycle)
 
 Suggested next action (subsequent milestone):
   Run /gsdd-new-milestone to start the next milestone cycle (gather goals, define requirements, create ROADMAP.md)
   Also available: /gsdd-progress (refresh after milestone setup)
 
 Suggested next action (incomplete milestone state — SPEC.md exists but no milestone archived yet):
-  Inspect .planning/ manually — a milestone is likely still in progress.
+  Inspect .work/ manually — a milestone is likely still in progress.
   If a ROADMAP.md was deleted prematurely, re-run /gsdd-new-milestone to restore it.
   Do NOT run /gsdd-new-project — SPEC.md already exists and re-running would overwrite it.
 
@@ -314,7 +315,7 @@ Handle compound states:
 - **Active brownfield change + non-matching `phase`/`quick` checkpoint:** Show the checkpoint as surviving context, but let the active brownfield change stay primary unless branch alignment, scope alignment, and still-active execution state all match.
 - **All phases complete + checkpoint:** All phases `[x]` but a checkpoint exists. If the checkpoint is `phase` or `quick`, mention both and suggest `/gsdd-resume` before continuing. If the checkpoint is `generic`, keep it visible as informational context and still route the primary recommendation to milestone audit.
 - **Phase done but next unplanned:** Current phase has both PLAN and SUMMARY, but the next phase has no PLAN. Show the current phase as complete and suggest planning the next phase (Branch C targeting the next phase).
-- **No matching condition:** If the project state does not match any branch, report it clearly and suggest the user inspect `.planning/` manually.
+- **No matching condition:** If the project state does not match any branch, report it clearly and suggest the user inspect `.work/` manually.
 </edge_cases>
 
 </process>

package/distilled/workflows/quick.md

@@ -16,13 +16,13 @@ They reuse the same planner, executor, and verifier roles but skip research and
 </anti_patterns>
 
 <prerequisites>
-`.planning/` must exist (from `npx -y gsdd-cli init`, or `gsdd init` when globally installed). ROADMAP.md is NOT required -- quick tasks work during any project phase.
+`.work/` must exist (from `npx -y gsdd-cli init`, or `gsdd init` when globally installed). ROADMAP.md is NOT required -- quick tasks work during any project phase.
 
-If `.planning/` does not exist, stop and tell the user to run `npx -y gsdd-cli init` first.
+If `.work/` does not exist, stop and tell the user to run `npx -y gsdd-cli init` first.
 </prerequisites>
 
 <repo_root_helper_contract>
-All `node .planning/bin/gsdd.mjs ...` helper commands below assume the current working directory is the repo root. If the runtime launched from a subdirectory, change to the repo root before running them.
+All `node .work/bin/gsdd.mjs ...` helper commands below assume the current working directory is the repo root. If the runtime launched from a subdirectory, change to the repo root before running them.
 </repo_root_helper_contract>
 
 <process>
@@ -37,17 +37,17 @@ Store the response as `$DESCRIPTION`. If empty, re-prompt.
 
 ## Step 2: Initialize
 
-1. Read `.planning/config.json` for workflow toggles and git protocol.
-2. Scan `.planning/quick/` for existing task directories. Calculate `$NEXT_NUM` as the next 3-digit number (001, 002, ...).
+1. Read `.work/config.json` for workflow toggles and git protocol.
+2. Scan `.work/quick/` for existing task directories. Calculate `$NEXT_NUM` as the next 3-digit number (001, 002, ...).
 3. Generate `$SLUG` from `$DESCRIPTION` (lowercase, hyphens, max 40 chars).
-4. Create `.planning/quick/$NEXT_NUM-$SLUG/`.
-5. If `.planning/brownfield-change/CHANGE.md` exists, read it first as the current bounded brownfield continuity anchor. Capture the active goal, current posture, next action, and declared write scope as `$BROWNFIELD_CONTEXT`. If `.planning/brownfield-change/HANDOFF.md` exists, read it as supporting judgment context only. Do not let `/gsdd-new-project` become the default fallback when this active change already defines a concrete bounded lane.
-6. If `.planning/codebase/` exists, read whichever of `.planning/codebase/ARCHITECTURE.md`, `.planning/codebase/STACK.md`, `.planning/codebase/CONVENTIONS.md`, and `.planning/codebase/CONCERNS.md` are present. Summarize key findings from available docs in <=500 words as `$CODEBASE_CONTEXT`, emphasizing: safest surfaces to touch, risky zones to avoid, must-know conventions/traps, and what must be re-verified after change. Note any missing docs in the summary.
-7. If `.planning/codebase/` does not exist, build a just-enough inline brownfield baseline instead of stopping. Read the repo root guidance that is cheap and stable (`README.md`, root manifest such as `package.json` / `pyproject.toml` / `Cargo.toml` when present, top-level app entrypoints, and any obviously relevant config or module files surfaced by `$DESCRIPTION`). Summarize the findings in <=500 words as `$CODEBASE_CONTEXT`, explicitly labeling it as a provisional baseline and calling out unknowns. Emphasize: likely implementation surface, likely dependency boundaries, conventions already visible, risky areas to avoid touching blindly, and what must be re-verified after the change. If the repo is still too unclear after this pass, keep that uncertainty explicit so Step 3.6 can recommend `/gsdd-map-codebase`.
-8. **Session-boundary fallback:** If `.planning/.continue-here.bak` exists, read its `<judgment>` section. Use `<active_constraints>` and `<anti_regression>` rules as task-scoping context (do not violate active constraints; do not regress on listed invariants). After reading, run `node .planning/bin/gsdd.mjs file-op delete .planning/.continue-here.bak --missing ok` (auto-clean).
-9. Inspect the live branch/worktree surface separately from checkpoint or planning artifacts. Run `node .planning/bin/gsdd.mjs control-map --json` when available and use its computed repo/worktree/planning truth to identify stale/spent branches, dirty tracked/untracked/ignored buckets, sibling or detached worktrees, local annotations, and cleanup obligations. This is advisory for quick tasks unless the mismatch makes the task description materially misleading; local annotations are intent hints, not product truth.
+4. Create `.work/quick/$NEXT_NUM-$SLUG/`.
+5. If `.work/brownfield-change/CHANGE.md` exists, read it first as the current bounded brownfield continuity anchor. Capture the active goal, current posture, next action, and declared write scope as `$BROWNFIELD_CONTEXT`. If `.work/brownfield-change/HANDOFF.md` exists, read it as supporting judgment context only. Do not let `/gsdd-new-project` become the default fallback when this active change already defines a concrete bounded lane.
+6. If `.work/codebase/` exists, read whichever of `.work/codebase/ARCHITECTURE.md`, `.work/codebase/STACK.md`, `.work/codebase/CONVENTIONS.md`, and `.work/codebase/CONCERNS.md` are present. Summarize key findings from available docs in <=500 words as `$CODEBASE_CONTEXT`, emphasizing: safest surfaces to touch, risky zones to avoid, must-know conventions/traps, and what must be re-verified after change. Note any missing docs in the summary.
+7. If `.work/codebase/` does not exist, build a just-enough inline brownfield baseline instead of stopping. Read the repo root guidance that is cheap and stable (`README.md`, root manifest such as `package.json` / `pyproject.toml` / `Cargo.toml` when present, top-level app entrypoints, and any obviously relevant config or module files surfaced by `$DESCRIPTION`). Summarize the findings in <=500 words as `$CODEBASE_CONTEXT`, explicitly labeling it as a provisional baseline and calling out unknowns. Emphasize: likely implementation surface, likely dependency boundaries, conventions already visible, risky areas to avoid touching blindly, and what must be re-verified after the change. If the repo is still too unclear after this pass, keep that uncertainty explicit so Step 3.6 can recommend `/gsdd-map-codebase`.
+8. **Session-boundary fallback:** If `.work/.continue-here.bak` exists, read its `<judgment>` section. Use `<active_constraints>` and `<anti_regression>` rules as task-scoping context (do not violate active constraints; do not regress on listed invariants). After reading, run `node .work/bin/gsdd.mjs file-op delete .work/.continue-here.bak --missing ok` (auto-clean).
+9. Inspect the live branch/worktree surface separately from checkpoint or planning artifacts. Run `node .work/bin/gsdd.mjs control-map --json` when available and use its computed repo/worktree/planning truth to identify stale/spent branches, dirty tracked/untracked/ignored buckets, sibling or detached worktrees, local annotations, and cleanup obligations. This is advisory for quick tasks unless the mismatch makes the task description materially misleading; local annotations are intent hints, not product truth.
 
-If `.planning/quick/` does not exist, create it along with an empty `LOG.md`:
+If `.work/quick/` does not exist, create it along with an empty `LOG.md`:
 
 ```markdown
 # Quick Task Log
@@ -60,7 +60,7 @@ If `.planning/quick/` does not exist, create it along with an empty `LOG.md`:
 
 ## Step 2.5: Approach clarification (conditional)
 
-Read `.planning/config.json`.
+Read `.work/config.json`.
 - If `workflow.discuss` is `false` (or key missing): set `$APPROACH_CONTEXT` to empty, skip to Step 3.
 - If `workflow.discuss` is `true`: evaluate `$DESCRIPTION` for ambiguity signals.
 
@@ -102,14 +102,14 @@ Delegate to the planner role in quick mode.
 
 <delegate>
 **Identity:** Planner (quick mode)
-**Instruction:** Read `.planning/templates/roles/planner.md` for your role contract, then create a plan for this quick task.
+**Instruction:** Read `.work/templates/roles/planner.md` for your role contract, then create a plan for this quick task.
 
 **Context to provide:**
 - Task description: `$DESCRIPTION`
 - Approach context: `$APPROACH_CONTEXT` (user-confirmed decisions from Step 2.5 — treat as locked constraints, do not revisit)
 - Codebase context: `$CODEBASE_CONTEXT` (existing brownfield codebase summary from codebase maps when they exist, otherwise the inline brownfield baseline from Step 2 — use for orientation and risk awareness, not as hard constraints)
 - Mode: quick (single plan, 1-3 tasks, no research phase)
-- Output path: `.planning/quick/$NEXT_NUM-$SLUG/$NEXT_NUM-PLAN.md`
+- Output path: `.work/quick/$NEXT_NUM-$SLUG/$NEXT_NUM-PLAN.md`
 
 **Constraints:**
 - If `$APPROACH_CONTEXT` is non-empty, implement the user's confirmed choices — do not substitute alternatives
@@ -118,21 +118,20 @@ Delegate to the planner role in quick mode.
 - No research phase, no ROADMAP requirements
 - Do NOT extract phase requirement IDs — there is no active phase
 - Derive must-haves directly from the task description
-- If the quick task is UI-sensitive, include proportional `ui_proof_slots` with slot_id, claim, route_state, required_evidence_kinds, minimum_observations, expected_artifact_types, validation_command, environment, viewport, manual_acceptance_required, and claim_limit; otherwise include a short `no_ui_proof_rationale`
-- UI proof slots must be matchable to exact observed evidence later: claim, route/state, observation, evidence kind, artifact path or manual step, privacy metadata, result, and claim limit. Discovery hints from source comments, AST/cAST, semantic search, or Semble-like retrieval do not satisfy proof.
-- Observed artifact metadata must include `visibility`, `retention`, `sensitivity`, and `safe_to_publish`; raw screenshots, traces, videos, DOM snapshots, and reports are local-only/unsafe by default. Use `gsdd ui-proof validate <path>` or `gsdd health` when a bundle exists; add `--claim <...>` only for public, publication, tracked, delivery, or release proof use.
+- If the quick task is UI-sensitive, include proportional `browser_proof_required: true`, a short `browser_proof_rationale`, and a plain Browser Proof Plan naming route/state, viewport, runtime path, evidence kind, evidence command or narrowed no-command rationale, observations, artifacts, and claim limit; otherwise set `browser_proof_required: false` with a short rationale.
+- Browser proof must be matchable to exact observed evidence later: route/state, viewport, observation, evidence kind, artifact path or manual step, privacy/safety note, result, and claim limit. Discovery hints from source comments, AST/cAST, semantic search, or Semble-like retrieval do not satisfy proof.
 - For live rendered UI proof, default to `agent-browser` snapshots/refs, interactions, screenshots, and relevant console/network observations. If unavailable, state the availability constraint and closest project-native interactive browser fallback before narrowing the claim. Existing Playwright/package-script browser tests remain the canonical repeatable regression path when present. The viewport set is plan-owned, but under-specified viewport coverage is weak proof; explain the chosen viewport(s) or narrow the claim limit.
-- Keep UI proof proportional: do not scaffold Playwright, Cypress, Cucumber, Storybook, CI, browser MCP, or visual-regression tooling by default
+- Keep browser proof proportional: do not scaffold Playwright, Cypress, Cucumber, Storybook, CI, browser MCP, or visual-regression tooling by default.
 - Ignore <planning_process> Step 1 requirement extraction; use inline goal-backward planning only
 - Target minimal context usage
 
-**Output:** `.planning/quick/$NEXT_NUM-$SLUG/$NEXT_NUM-PLAN.md`
+**Output:** `.work/quick/$NEXT_NUM-$SLUG/$NEXT_NUM-PLAN.md`
 **Return:** Plan file path and task count.
 </delegate>
 
 After the planner returns:
 
-**STOP. Verify that `.planning/quick/$NEXT_NUM-$SLUG/$NEXT_NUM-PLAN.md` exists on disk before proceeding to execution. If the file does not exist, report the error to the user and do NOT proceed. A plan that exists only in conversation context will be lost.**
+**STOP. Verify that `.work/quick/$NEXT_NUM-$SLUG/$NEXT_NUM-PLAN.md` exists on disk before proceeding to execution. If the file does not exist, report the error to the user and do NOT proceed. A plan that exists only in conversation context will be lost.**
 
 ### Quick Plan Self-Check
 
@@ -147,23 +146,23 @@ This is a self-check, not an independent plan-check. Failures are noted but do N
 
 ## Step 3.5: Independent plan check (conditional)
 
-Read `.planning/config.json`.
+Read `.work/config.json`.
 - If `workflow.planCheck` is `false` (or key missing): skip to Step 3.6.
 - If `workflow.planCheck` is `true`: delegate to the plan-checker with quick-scoped dimensions.
 
 <delegate>
 **Identity:** Plan Checker (quick mode)
-**Instruction:** Read `.planning/templates/delegates/plan-checker.md` for your role contract, then check this quick task plan.
+**Instruction:** Read `.work/templates/delegates/plan-checker.md` for your role contract, then check this quick task plan.
 
 **Context to provide:**
 - Task description: `$DESCRIPTION` (treat as the phase goal equivalent)
-- Plan: `.planning/quick/$NEXT_NUM-$SLUG/$NEXT_NUM-PLAN.md`
+- Plan: `.work/quick/$NEXT_NUM-$SLUG/$NEXT_NUM-PLAN.md`
 - Mode: quick
 
 **Constraints:**
 - Check 5 dimensions by default: `requirement_coverage`, `task_completeness`, `dependency_correctness`, `scope_sanity`, `must_have_quality`
-- If the quick plan contains `ui_proof_slots` or a rendered UI claim, also check `closure_honesty` so weak UI proof slots block execution
-- Skip: `key_link_completeness`, `context_compliance`, `goal_achievement`, `approach_alignment`
+- If the quick plan contains a rendered UI claim, check `goal_achievement` so weak browser proof blocks execution.
+- Skip: `key_link_completeness`, `context_compliance`, and `approach_alignment`; skip `goal_achievement` only when the quick plan has no rendered UI claim
 - Maximum 1 revision cycle (if blockers found, send back to planner once, then accept result)
 - Blocker threshold: only block on `task_completeness` or `scope_sanity` violations
 - Warnings for other dimensions are noted but do not block
@@ -190,8 +189,8 @@ Evaluate the plan against quick-scope boundaries. Read the plan file and check:
 | Files modified | >8 distinct files in plan | "This task touches {N} files — consider `/gsdd-plan` for full ceremony." |
 | Architecture keywords in `$DESCRIPTION` | contains: `refactor`, `migration`, `security`, `auth`, `API design`, `schema`, `database` | "This looks like architectural work — consider `/gsdd-plan` for approach exploration." |
 | New public APIs | Plan tasks create new route files, API endpoints, or exported interfaces | "New public surface area detected — consider `/gsdd-plan` for approach exploration." |
-| Orientation gap | No `.planning/codebase/` exists AND the inline brownfield baseline still cannot name a clear implementation surface, dependency boundary, or safe-to-touch module | "This repo still needs deeper orientation — consider `/gsdd-map-codebase` before changing code." |
-| Undefined bounded change | `$DESCRIPTION` still does not identify a concrete bug, feature, target surface, or observable outcome after clarification | "This does not yet describe a bounded change — use `/gsdd-new-project` to define the work first. If `.planning/brownfield-change/CHANGE.md` already defines a concrete bounded lane, treat `/gsdd-new-project` as an intentional widen path rather than the default fallback." |
+| Orientation gap | No `.work/codebase/` exists AND the inline brownfield baseline still cannot name a clear implementation surface, dependency boundary, or safe-to-touch module | "This repo still needs deeper orientation — consider `/gsdd-map-codebase` before changing code." |
+| Undefined bounded change | `$DESCRIPTION` still does not identify a concrete bug, feature, target surface, or observable outcome after clarification | "This does not yet describe a bounded change — use `/gsdd-new-project` to define the work first. If `.work/brownfield-change/CHANGE.md` already defines a concrete bounded lane, treat `/gsdd-new-project` as an intentional widen path rather than the default fallback." |
 
 If any signals fire, concatenate the matching advisory text in the listed order as `$SCOPE_WARNING`. If the undefined bounded change signal fires, keep that advisory first so the routing recommendation stays explicit.
 If no signals fire, `$SCOPE_WARNING` is empty.
@@ -256,11 +255,11 @@ Delegate to the executor role.
 
 <delegate>
 **Identity:** Executor
-**Instruction:** Read `.planning/templates/roles/executor.md` for your role contract, then execute the quick task plan.
+**Instruction:** Read `.work/templates/roles/executor.md` for your role contract, then execute the quick task plan.
 
 **Context to provide:**
-- Plan file: `.planning/quick/$NEXT_NUM-$SLUG/$NEXT_NUM-PLAN.md`
-- Project conventions: `.planning/config.json` (git protocol section)
+- Plan file: `.work/quick/$NEXT_NUM-$SLUG/$NEXT_NUM-PLAN.md`
+- Project conventions: `.work/config.json` (git protocol section)
 - Quick task -- do NOT update ROADMAP.md
 
 **Constraints:**
@@ -268,52 +267,52 @@ Delegate to the executor role.
 - Follow advisory git protocol from config.json
 - Skip the <state_updates> section of your role contract entirely
 - Do NOT update ROADMAP.md phase status or SPEC.md current state
-- Create summary at: `.planning/quick/$NEXT_NUM-$SLUG/$NEXT_NUM-SUMMARY.md`
-- If the quick plan defines `ui_proof_slots`, create or update `.planning/quick/$NEXT_NUM-$SLUG/UI-PROOF.md` with fenced JSON containing required top-level fields: `proof_bundle_version`, `scope`, `route_state`, `environment`, `viewport`, `evidence_inputs`, `commands_or_manual_steps`, `observations`, `artifacts`, `privacy`, `result`, and `claim_limits`
-- For live UI proof, record `agent-browser` in `evidence_inputs.tools_used` when used, the exact commands or manual ref-based steps, screenshot/report artifact paths, and any relevant console/network observations. If `agent-browser` was unavailable, record that availability constraint and fallback tool explicitly. If existing Playwright tests supplied regression evidence, record the package command and result separately from the `agent-browser` runtime observation.
+- Create summary at: `.work/quick/$NEXT_NUM-$SLUG/$NEXT_NUM-SUMMARY.md`
+- If the quick plan requires browser proof, create or update `.work/quick/$NEXT_NUM-$SLUG/UI-PROOF.md` as a plain observation record with route/state, viewport, runtime path, evidence command or narrowed no-command rationale, observations, artifacts, privacy/safety note, result, and claim limit.
+- For live browser proof, record `agent-browser` when used, the exact commands or manual ref-based steps, screenshot/report artifact paths, and any relevant console/network observations. If `agent-browser` was unavailable, record that availability constraint and fallback tool explicitly. If existing Playwright tests supplied regression evidence, record the package command and result separately from the runtime observation.
 - Human approval for visual taste, accessibility judgment, baseline acceptance, subjective polish/layout quality, or privacy publication does not replace required `code`, `test`, `runtime`, or `delivery` evidence
 
-**Output:** `.planning/quick/$NEXT_NUM-$SLUG/$NEXT_NUM-SUMMARY.md`
+**Output:** `.work/quick/$NEXT_NUM-$SLUG/$NEXT_NUM-SUMMARY.md`
 **Return:** Summary file path and completion status.
 </delegate>
 
 After the executor returns:
 
-**STOP. Verify the SUMMARY file exists at `.planning/quick/$NEXT_NUM-$SLUG/$NEXT_NUM-SUMMARY.md` on disk. If it does not exist, report the write failure. Do NOT proceed to verification or LOG.md update without a persisted summary.**
+**STOP. Verify the SUMMARY file exists at `.work/quick/$NEXT_NUM-$SLUG/$NEXT_NUM-SUMMARY.md` on disk. If it does not exist, report the write failure. Do NOT proceed to verification or LOG.md update without a persisted summary.**
 
 ---
 
 ## Step 5: Verify (conditional)
 
-Read `.planning/config.json`.
+Read `.work/config.json`.
 - If `workflow.verifier` is `false`, skip to Step 6.
 - If `workflow.verifier` is `true`, delegate to the verifier role:
 
 <delegate>
 **Identity:** Verifier (quick mode)
-**Instruction:** Read `.planning/templates/roles/verifier.md` for your role contract, then verify the quick task.
+**Instruction:** Read `.work/templates/roles/verifier.md` for your role contract, then verify the quick task.
 
 **Context to provide:**
 - Task description: `$DESCRIPTION`
-- Plan: `.planning/quick/$NEXT_NUM-$SLUG/$NEXT_NUM-PLAN.md`
-- Summary: `.planning/quick/$NEXT_NUM-$SLUG/$NEXT_NUM-SUMMARY.md`
+- Plan: `.work/quick/$NEXT_NUM-$SLUG/$NEXT_NUM-PLAN.md`
+- Summary: `.work/quick/$NEXT_NUM-$SLUG/$NEXT_NUM-SUMMARY.md`
 
 **Constraints:**
 - Verify goal achievement against the task description
 - Quick scope -- do not check ROADMAP alignment or cross-phase integration
-- Write report to: `.planning/quick/$NEXT_NUM-$SLUG/$NEXT_NUM-VERIFICATION.md`
+- Write report to: `.work/quick/$NEXT_NUM-$SLUG/$NEXT_NUM-VERIFICATION.md`
 
-**Output:** `.planning/quick/$NEXT_NUM-$SLUG/$NEXT_NUM-VERIFICATION.md`
+**Output:** `.work/quick/$NEXT_NUM-$SLUG/$NEXT_NUM-VERIFICATION.md`
 **Return:** Verification status (passed | gaps_found | human_needed).
 </delegate>
 
-**STOP. Verify the VERIFICATION file exists at `.planning/quick/$NEXT_NUM-$SLUG/$NEXT_NUM-VERIFICATION.md` on disk (when verifier ran). If it does not exist, report the write failure. Do NOT proceed to LOG.md update without a persisted verification report.**
+**STOP. Verify the VERIFICATION file exists at `.work/quick/$NEXT_NUM-$SLUG/$NEXT_NUM-VERIFICATION.md` on disk (when verifier ran). If it does not exist, report the write failure. Do NOT proceed to LOG.md update without a persisted verification report.**
 
 ---
 
 ## Step 6: Update LOG.md
 
-Append a row to `.planning/quick/LOG.md`:
+Append a row to `.work/quick/LOG.md`:
 
 ```markdown
 | $NEXT_NUM | $DESCRIPTION | $DATE | $STATUS | [$NEXT_NUM-$SLUG](./$NEXT_NUM-$SLUG/) |
@@ -339,8 +338,8 @@ Report to the user:
 <success_criteria>
 - [ ] User provided a task description
 - [ ] Approach clarification ran (only if workflow.discuss is true AND ambiguity detected)
-- [ ] `.planning/quick/` directory exists (created if needed)
-- [ ] Task directory created at `.planning/quick/NNN-slug/`
+- [ ] `.work/quick/` directory exists (created if needed)
+- [ ] Task directory created at `.work/quick/NNN-slug/`
 - [ ] `NNN-PLAN.md` created by planner (1-3 tasks)
 - [ ] Independent plan check ran (only if workflow.planCheck is true)
 - [ ] Plan preview presented to user before execution
@@ -358,10 +357,10 @@ Report to the user what was accomplished, then present the next step:
 **Completed:** Quick task #{next_num} — {description}
 
 Created:
-- `.planning/quick/{next_num}-{slug}/{next_num}-PLAN.md`
-- `.planning/quick/{next_num}-{slug}/{next_num}-SUMMARY.md`
-- `.planning/quick/{next_num}-{slug}/{next_num}-VERIFICATION.md` (if verifier enabled)
-- Updated `.planning/quick/LOG.md`
+- `.work/quick/{next_num}-{slug}/{next_num}-PLAN.md`
+- `.work/quick/{next_num}-{slug}/{next_num}-SUMMARY.md`
+- `.work/quick/{next_num}-{slug}/{next_num}-VERIFICATION.md` (if verifier enabled)
+- Updated `.work/quick/LOG.md`
 
 **Next step:** `/gsdd-progress` — check project status and continue phase work
 

package/distilled/workflows/resume.md

@@ -7,15 +7,15 @@ Scope boundary: unlike progress.md, you have side effects — checkpoint cleanup
 </role>
 
 <prerequisites>
-`.planning/` should exist. If it does not, route the user to `npx -y gsdd-cli init`.
+`.work/` should exist. If it does not, route the user to `npx -y gsdd-cli init`.
 </prerequisites>
 
 <repo_root_helper_contract>
-All `node .planning/bin/gsdd.mjs ...` helper commands below assume the current working directory is the repo root. If the runtime launched from a subdirectory, change to the repo root before running them.
+All `node .work/bin/gsdd.mjs ...` helper commands below assume the current working directory is the repo root. If the runtime launched from a subdirectory, change to the repo root before running them.
 </repo_root_helper_contract>
 
 <runtime_contract>
-Use the `Runtime` type from `.planning/SPEC.md`.
+Use the `Runtime` type from `.work/SPEC.md`.
 Infer runtime from the launching surface when obvious: `.claude/` -> `claude-code`, `.codex/` or Codex portable skill -> `codex-cli`, `.opencode/` -> `opencode`, otherwise `other`.
 When a checkpoint's `runtime` differs from the inferred current runtime, surface it as an informational note in `<present_status>` — it is context, not a gate.
 </runtime_contract>
@@ -23,7 +23,7 @@ When a checkpoint's `runtime` differs from the inferred current runtime, surface
 <lifecycle_preflight>
 Before loading checkpoint state or cleaning up any checkpoint file, run:
 
-- `node .planning/bin/gsdd.mjs lifecycle-preflight resume`
+- `node .work/bin/gsdd.mjs lifecycle-preflight resume`
 
 If the preflight result is `blocked`, STOP and report the blocker instead of inferring resume eligibility from workflow-local prose.
 
@@ -36,15 +36,16 @@ Treat the preflight as an authorization seam over shared repo truth only:
 <process>
 
 <control_map_reconciliation>
-Before routing from a checkpoint, run `node .planning/bin/gsdd.mjs control-map --json` when available. Reconcile the checkpoint narrative against computed repo/worktree/planning truth: canonical branch/HEAD, tracked/untracked/ignored dirty buckets, sibling or detached worktrees, local annotations, active brownfield anchors, and planning drift. If the checkpoint understates dirty work, points at a stale branch/worktree, or conflicts with the active planning/brownfield surface, stop and present the mismatch before recommending execution. Local annotations are useful intent hints, not authority.
+Before routing from a checkpoint, run `node .work/bin/gsdd.mjs control-map --json` when available. Reconcile the checkpoint narrative against computed repo/worktree/planning truth: canonical branch/HEAD, tracked/untracked/ignored dirty buckets, sibling or detached worktrees, local annotations, active brownfield anchors, and planning drift. If the checkpoint understates dirty work, points at a stale branch/worktree, or conflicts with the active planning/brownfield surface, stop and present the mismatch before recommending execution. Local annotations are useful intent hints, not authority.
 </control_map_reconciliation>
 
 <detect_state>
 Check for project artifacts in order:
 
-1. **No `.planning/` directory** — route user to run `npx -y gsdd-cli init`. Stop.
-2. **If `.planning/brownfield-change/CHANGE.md` exists** — this repo has an active medium-scope brownfield change. Proceed to load brownfield continuity state even if there is no active roadmap.
-3. **No `.planning/SPEC.md` or no `.planning/ROADMAP.md`** — `.planning/` exists but the project is not fully initialized (partial init). Route user to run the `/gsdd-new-project` workflow. Stop.
+1. **No `.work/` directory** — route user to run `npx -y gsdd-cli init`. Stop.
+2. **If `.work/brownfield-change/CHANGE.md` exists and `Current posture` is not `closed`** — this repo has an active medium-scope brownfield change. Proceed to load brownfield continuity state even if there is no active roadmap.
+   - If `Current posture` is `closed`, treat the file as historical context only; it does not replace a checkpoint or active roadmap.
+3. **No `.work/SPEC.md` or no `.work/ROADMAP.md`** — `.work/` exists but the project is not fully initialized (partial init). Route user to run the `/gsdd-new-project` workflow. Stop.
 4. **Both exist** — proceed to load state.
 </detect_state>
 
@@ -52,7 +53,7 @@ Check for project artifacts in order:
 Read the following files and extract state:
 
 **ROADMAP.md:**
-If `.planning/ROADMAP.md` exists, read it. If it is absent because the active brownfield change is the only continuity anchor, keep phase fields empty and continue from `CHANGE.md`.
+If `.work/ROADMAP.md` exists, read it. If it is absent because the active brownfield change is the only continuity anchor, keep phase fields empty and continue from `CHANGE.md`.
 
 When present, parse phase statuses:
 - `[ ]` = not started
@@ -66,14 +67,14 @@ Determine:
 - Completed phase count
 
 **SPEC.md:**
-If `.planning/SPEC.md` exists, read it. If it is absent because the active brownfield change is the only continuity anchor, label project identity as `unknown from SPEC.md` and derive runtime/context from the launching surface plus `CHANGE.md` / `HANDOFF.md`.
+If `.work/SPEC.md` exists, read it. If it is absent because the active brownfield change is the only continuity anchor, label project identity as `unknown from SPEC.md` and derive runtime/context from the launching surface plus `CHANGE.md` / `HANDOFF.md`.
 
 When present, extract:
 - Project name or description (first heading or "What This Is" section)
 - Current state summary if present
 
 **Active brownfield change:**
-If `.planning/brownfield-change/CHANGE.md` exists, read it first as the canonical operational continuity anchor and extract:
+If `.work/brownfield-change/CHANGE.md` exists and is not closed, read it first as the canonical operational continuity anchor and extract:
 - change title from the first heading
 - current posture from `## Current Status`
 - current branch / integration surface from `## Current Status`
@@ -81,7 +82,7 @@ If `.planning/brownfield-change/CHANGE.md` exists, read it first as the canonica
 - next action from `## Next Action`
 - declared write scope from `## PR Slice Ownership` when present
 
-If `.planning/brownfield-change/HANDOFF.md` exists, read it as judgment-only context:
+If `.work/brownfield-change/HANDOFF.md` exists, read it as judgment-only context:
 - active constraints
 - unresolved uncertainty
 - decision posture
@@ -91,7 +92,7 @@ If `.planning/brownfield-change/HANDOFF.md` exists, read it as judgment-only con
 Do not flatten `CHANGE.md` and `HANDOFF.md` into co-equal operational sources. `CHANGE.md` stays the live status/next-action anchor; `HANDOFF.md` explains why that posture exists.
 
 **Checkpoint file:**
-Check if `.planning/.continue-here.md` exists. If yes, read it and extract:
+Check if `.work/.continue-here.md` exists. If yes, read it and extract:
 - `workflow` frontmatter (phase/quick/generic)
 - `phase` frontmatter
 - `runtime` frontmatter (the runtime that wrote the checkpoint; use `unknown` if field absent)
@@ -99,12 +100,12 @@ Check if `.planning/.continue-here.md` exists. If yes, read it and extract:
 - `<judgment>` if present, including `<active_constraints>`, `<unresolved_uncertainty>`, `<decision_posture>`, and `<anti_regression>`
 
 **Phase directories:**
-Scan `.planning/phases/` for:
+Scan `.work/phases/` for:
 - Directories with a PLAN file but no SUMMARY file (incomplete execution)
-- Directories with a SUMMARY file but no VERIFICATION file (unverified phase, if `workflow.verifier` is enabled in `.planning/config.json`; if config.json cannot be read, assume verifier is disabled)
+- Directories with a SUMMARY file but no VERIFICATION file (unverified phase, if `workflow.verifier` is enabled in `.work/config.json`; if config.json cannot be read, assume verifier is disabled)
 
 **Quick task log:**
-If `.planning/quick/LOG.md` exists, read the last entry. Check if it has a non-terminal status (not `done`/`passed`).
+If `.work/quick/LOG.md` exists, read the last entry. Check if it has a non-terminal status (not `done`/`passed`).
 
 **Git/worktree truth:**
 Collect the live integration-surface facts separately from checkpoint narrative truth:
@@ -120,7 +121,7 @@ Collect the live integration-surface facts separately from checkpoint narrative
 <provenance_reconciliation>
 Before routing, reconstruct and compare these truth buckets explicitly:
 
-1. **Checkpoint narrative truth** — what `.planning/.continue-here.md` claims was happening
+1. **Checkpoint narrative truth** — what `.work/.continue-here.md` claims was happening
 2. **Planning/artifact truth** — what ROADMAP, SPEC, phase files, quick-task logs, and the active brownfield change artifacts say
 3. **Git/worktree truth** — what the live branch and working tree say now
 
@@ -152,12 +153,12 @@ If git/worktree truth materially disagrees with the active brownfield artifact:
 </provenance_reconciliation>
 
 <validate_checkpoint>
-Only run this step when `.planning/.continue-here.md` was found in `<load_artifacts>`. If no checkpoint exists, skip this step entirely.
+Only run this step when `.work/.continue-here.md` was found in `<load_artifacts>`. If no checkpoint exists, skip this step entirely.
 
 Cross-validate checkpoint fields against current roadmap state in this order:
 
 1. Extract the checkpoint `workflow` and `phase` frontmatter fields.
-2. If `phase` is non-null, look up that exact phase name in `.planning/ROADMAP.md`.
+2. If `phase` is non-null, look up that exact phase name in `.work/ROADMAP.md`.
    - If the matching roadmap entry is `[x]`, mark the checkpoint as stale.
    - Record the specific reason in this form: `checkpoint references phase "[phase]" which is already complete [x] in ROADMAP.md`.
    - Stop further staleness checks after recording this reason.
@@ -280,7 +281,7 @@ If `<validate_checkpoint>` marked the checkpoint as stale, keep the same routing
 If `<provenance_reconciliation>` marked a material checkpoint/worktree mismatch, keep the same routing logic but require explicit acknowledgement before continuing. The workflow should not silently route onward from a materially misleading checkpoint narrative.
 If an active brownfield change also exists, apply the same strict-match rule before making a `phase` or `quick` checkpoint primary. A checkpoint only stays primary when branch alignment, scope alignment, and still-active execution state all hold at once. Otherwise, keep the checkpoint visible and user-selectable, but fall through to the active brownfield change as the default resume target.
 
-**Active brownfield change (`.planning/brownfield-change/CHANGE.md`):**
+**Active brownfield change (`.work/brownfield-change/CHANGE.md`):**
 If there is no checkpoint, or if the surviving checkpoint does not satisfy the strict-match rule against the active brownfield change:
 - present the `CHANGE.md` next action as the primary resume target
 - keep `HANDOFF.md` as supporting judgment context only
@@ -327,8 +328,8 @@ Wait for user selection.
 <cleanup_checkpoint>
 Immediately after the user confirms their action selection (before routing to the target workflow):
 - If the user chose to resume from `.continue-here.md`:
-  1. Run `node .planning/bin/gsdd.mjs file-op copy .planning/.continue-here.md .planning/.continue-here.bak`.
-  2. After the copy succeeds, run `node .planning/bin/gsdd.mjs file-op delete .planning/.continue-here.md`.
+  1. Run `node .work/bin/gsdd.mjs file-op copy .work/.continue-here.md .work/.continue-here.bak`.
+  2. After the copy succeeds, run `node .work/bin/gsdd.mjs file-op delete .work/.continue-here.md`.
 - If the user chose a different action (not based on the checkpoint), leave `.continue-here.md` in place for a future resume.
 
 Copying before deleting ensures the checkpoint survives a session crash between deletion and dispatch. `.continue-here.bak` is cleaned up by the downstream workflow after absorbing the judgment, or by the next `pause.md` run.