@gobing-ai/spur 0.3.78 → 0.3.81

package/config/transition-shims.json

@@ -15,19 +15,19 @@
             "keepsWorking": "a pre-existing agent spec carrying only type (no executor field) still drains: --agent <specId> falls back to the spec's coding-agent type instead of resolving a missing executor",
             "removalCondition": "every spec under .spur/agents/ records an executor field (scan .spur/agents/*.yaml for the key; re-materialize the demo team first)"
         },
-        {
-            "id": "agent-flag-spec-id",
-            "wbs": "0542",
-            "file": "apps/cli/src/commands/agent.ts",
-            "keepsWorking": "a team spec id passed to --agent <spec-id> still addresses the spec and drives the drain path, warned once by warnAgentSpecIdOnce; --spec <id> is the canonical carrier",
-            "removalCondition": "no --agent <spec-id> usage remains in config/workflows/, plugins/sp/, or docs/ (scan for --agent values that match a .spur/agents/ spec id)"
-        },
         {
             "id": "agent-default-executor",
             "wbs": "0542",
             "file": "packages/app/src/services/agent-service.ts",
             "keepsWorking": "a configured executor name in agent.default still resolves during the transition, warned once by warnAgentDefaultExecutorOnce; the value domain moved to roles (recommended default: coder)",
             "removalCondition": "no agent.default value names an agent.executors entry (scan .spur/config.yaml and config/config.example.yaml against the agent.executors names)"
+        },
+        {
+            "id": "team-noun-retired",
+            "wbs": "0848",
+            "file": "apps/cli/src/commands/team.ts",
+            "keepsWorking": "the `spur team` noun keeps working after its six verbs moved to owning nouns (0848): assign → `spur task update --assignee`, status → `spur agent list --specs`, up → fleet materialization at serve start (`spur projects list --fleet` for --check), down/start/stop → `spur agent stop|start|stop <spec-id>`; a one-time stderr warning (warnTeamNounRetiredOnce) names the replacement for the invoked verb and every verb keeps its previous exit code and output",
+            "removalCondition": "no `spur team` invocation remains in config/workflows/, plugins/sp/, scripts/, or docs/, and the cutover window is recorded in docs/features/G64_retire-workspace-inbox-teams-and-spur-team.md"
         }
     ]
 }

package/config/workflows/basic.yaml

@@ -56,6 +56,8 @@ states:
     onEnter:
       - kind: shell
         options:
+          # (e) 0825: 10 lines — soft probe must run the trusted compound gate via `sh -c`
+          # and write the status file the guards read, always exit 0 (0771 R1 contract).
           command: >-
             mkdir -p .spur/run &&
             STATUS_FILE=".spur/run/$__runId-basic-gate.status" &&
@@ -83,6 +85,8 @@ states:
         options:
           agent: ${vars.agent}
           input: /sp:dev-fixall "${vars.qualityGateCmd}"
+          answerFile: .spur/run/${vars.__runId}-basic-fix-answer.txt
+          expectFile: .spur/run/${vars.__runId}-basic-fix-answer.txt
           # Declared Layer-1 role (0538 R2): routing reason beside the agent: pin.
           role: coder
           timeoutMs: ${vars.stepTimeoutMs}

package/config/workflows/docs-pipeline.yaml

@@ -138,12 +138,8 @@ states:
       # excluded from the digest's git-tree half, so the spec is folded in via `taskFile`.
       - kind: shell
         options:
-          # 0760 R1 (sibling of 0751 R2): task-path lookup fails closed. Drop
-          # `2>/dev/null`, `|| true`, and the forced `exit 0` so an unresolved
-          # wbs exits non-zero with a named message. The resolved path folds
-          # into the proof digest via `taskFile:` below; a silent miss would
-          # degrade the docs-pipeline proof to tree-only (same hazard 0751 R2
-          # removed from task-pipeline). (0769: run-scoped capture path.)
+          # (e) 0825: 6 lines — fail-closed lookup feeds the verdict/digest file names
+          # (0760 R1); a silent miss would degrade the proof to tree-only.
           command: >-
             set -e;
             task_path="$($spurBin task path $wbs --json | jq -r '.path // .filePath // empty')";
@@ -159,6 +155,8 @@ states:
       # folding a stale spec into the proof.
       - kind: shell
         options:
+          # (e) 0825: 9 lines — per-check rc capture (task show/feature show/jq) keeps an
+          # orphan task and a broken linked feature distinguishable (0785 R2 contract).
           command: >-
             set -e;
             FID="$($spurBin task show $wbs --json 2>/dev/null | jq -r '.feature_id // .frontmatter.feature_id // empty' 2>/dev/null)";
@@ -191,6 +189,7 @@ states:
           role: reviewer
           timeoutMs: ${vars.stepTimeoutMs}
           answerFile: .spur/run/${vars.__runId}-docs-verify-answer.txt
+          expectFile: .spur/run/${vars.__runId}-docs-verify-answer.txt
       - kind: shell
         options:
           command: "$spurBin task verdict $wbs --from-answer .spur/run/$__runId-docs-verify-answer.txt"
@@ -199,6 +198,8 @@ states:
       # guard below, not this step.
       - kind: shell
         options:
+          # (e) 0825: 6 lines — stamp shell freezes the docs proof digest for the closing
+          # guard; the record→done guard compares against this stamp.
           command: >-
             V=".spur/run/$wbs-verdict.json";
             if [ -f "$V" ] && [ -n "$proofDigest" ]; then
@@ -263,7 +264,7 @@ transitions:
     guard:
       kind: shell
       options:
-        command: 'test "$(cat .spur/run/$wbs-docs-precheck.status 2>/dev/null)" = PASS'
+        command: 'test "$(cat .spur/run/$__runId-docs-precheck.status 2>/dev/null)" = PASS'
   - from: precheck
     to: failed
     description: Precheck FAIL — stop before draft.
@@ -313,17 +314,13 @@ transitions:
     guard:
       kind: shell
       options:
-        # Hardened (0703 P2 carry): every operand is required non-empty (`test -n`) so a
-        # missing file or empty var fails closed instead of empty-equals-empty passing.
+        # (e) 0825: 8-line operand ladder condensed to one jq select — same file, same
+        # digest-locked PASS semantics (0769 R4/R5); empty/malformed still fail closed.
         command: >-
           test -n "$proofDigest" &&
           test -n "$proofDigestNow" &&
           test "$proofDigestNow" = "$proofDigest" &&
-          test -f ".spur/run/$wbs-verdict.json" &&
-          test -n "$(jq -r '.verdict // empty' .spur/run/$wbs-verdict.json 2>/dev/null)" &&
-          test "$(jq -r '.verdict // empty' .spur/run/$wbs-verdict.json 2>/dev/null)" = PASS &&
-          test -n "$(jq -r '.proof.digest // empty' .spur/run/$wbs-verdict.json 2>/dev/null)" &&
-          test "$(jq -r '.proof.digest // empty' .spur/run/$wbs-verdict.json 2>/dev/null)" = "$proofDigest"
+          test "$(jq -r --arg d "$proofDigest" 'select(.proof.digest == $d) | .verdict // empty' ".spur/run/$wbs-verdict.json" 2>/dev/null)" = PASS
   - from: verify
     to: failed
     description: Non-PASS verdict, missing/malformed answer, or digest mismatch (R5) — the run
@@ -338,6 +335,8 @@ transitions:
     guard:
       kind: shell
       options:
+        # (e) 0825: 5-line guard — captured record PASS + digest-locked persisted verdict
+        # must both hold before done; a failed `task record` never converts to success.
         command: >-
           test "$(cat .spur/run/$__runId-docs-record.status 2>/dev/null)" = PASS &&
           test -f ".spur/run/$wbs-verdict.json" &&

package/config/workflows/feature-dev.yaml

@@ -91,35 +91,11 @@ states:
     onEnter:
       - kind: shell
         options:
+          # (d) 0825: the 12-command identity/roster program moved into
+          # plugins/sp/scripts/feature-dev-precheck.ts (§1.1 owner); same messages,
+          # artifacts and soft-fail exit-0 contract, wrapper fails closed.
           command: >-
-            mkdir -p .spur/run;
-            FEATURE_JSON=".spur/run/$__runId-feature-dev-feature.json";
-            ROSTER_JSON=".spur/run/$__runId-feature-dev-roster.json";
-            TASKS_TXT=".spur/run/$__runId-feature-dev-tasks.txt";
-            PRECHECK_STATUS=".spur/run/$__runId-feature-dev-precheck.status";
-            rm -f "$PRECHECK_STATUS";
-            set +e;
-            test -n "$featureId" && test -n "$__runId";
-            id_rc=$?;
-            if [ "$id_rc" -eq 0 ]; then $spurBin feature show "$featureId" --json > "$FEATURE_JSON" 2>&1; show_rc=$?; else show_rc=1; fi;
-            if [ "$id_rc" -eq 0 ] && [ "$show_rc" -eq 0 ] && jq -e --arg id "$featureId" 'type == "object" and .id == $id' "$FEATURE_JSON" > /dev/null 2>&1; then $spurBin task list --feature "$featureId" --json > "$ROSTER_JSON" 2>&1; list_rc=$?; else list_rc=1; fi;
-            if [ "$id_rc" -ne 0 ] || [ "$show_rc" -ne 0 ] || [ "$list_rc" -ne 0 ]; then
-              echo "feature-dev precheck: missing featureId/runId, unknown feature '$featureId', or unreadable roster (rc $id_rc/$show_rc/$list_rc) — supply an existing planned feature via /sp:dev-plan or /sp:dev-idea; nothing was auto-created or re-planned" >&2;
-              printf 'FAIL\n' > "$PRECHECK_STATUS";
-            elif ! jq -e 'type == "array" and length > 0' "$ROSTER_JSON" > /dev/null 2>&1; then
-              echo "feature-dev precheck: roster at $ROSTER_JSON is malformed, not an array, or empty — plan the feature first via /sp:dev-plan; refusing to replan or run an empty batch" >&2;
-              printf 'FAIL\n' > "$PRECHECK_STATUS";
-            elif ! jq -e 'all(.[]; (.wbs | type == "string" and length > 0)) and ([.[].wbs] | length == (unique | length)) and all(.[]; .status == "todo" or .status == "done" or .status == "cancelled" or .status == "backlog" or .status == "wip" or .status == "testing" or .status == "blocked")' "$ROSTER_JSON" > /dev/null 2>&1; then
-              echo "feature-dev precheck: roster has empty/duplicate/mismatched WBS identities or unknown statuses at $ROSTER_JSON — repair the task corpus; refusing to batch a broken roster" >&2;
-              printf 'FAIL\n' > "$PRECHECK_STATUS";
-            elif [ "$(jq '[.[] | select(.status == "backlog" or .status == "wip" or .status == "testing" or .status == "blocked")] | length' "$ROSTER_JSON")" -gt 0 ]; then
-              echo "feature-dev precheck: linked task(s) are backlog/wip/testing/blocked — refine or resume them through their own task pipelines before batching; refusing to launch overlapping work" >&2;
-              printf 'FAIL\n' > "$PRECHECK_STATUS";
-            else
-              printf '%s' "$(jq -r '[.[] | select(.status == "todo") | .wbs] | sort | join(",")' "$ROSTER_JSON")" > "$TASKS_TXT";
-              printf 'PASS\n' > "$PRECHECK_STATUS";
-            fi;
-            exit 0
+            mkdir -p .spur/run; if [ -f plugins/sp/scripts/feature-dev-precheck.ts ]; then bun plugins/sp/scripts/feature-dev-precheck.ts; elif W="$(superskill script path sp feature-dev-precheck.mjs 2>/dev/null)" && [ -f "$W" ]; then node "$W"; else echo "feature-dev precheck failed closed — feature-dev-precheck script not found — run 'superskill install sp'" >&2; printf 'FAIL\n' > ".spur/run/$__runId-feature-dev-precheck.status"; fi; exit 0
       - kind: note
         options:
           message: "Feature-dev start for existing feature ${vars.featureId} (0782 reuse path): roster resolved under .spur/run/${vars.__runId}-feature-dev-roster.json; frozen todo list at .spur/run/${vars.__runId}-feature-dev-tasks.txt."
@@ -143,6 +119,8 @@ states:
         options:
           agent: ${vars.agent}
           input: /sp:dev-runall --tasks ${vars.featureTaskIds} --auto
+          answerFile: .spur/run/${vars.__runId}-feature-dev-runall-answer.txt
+          expectFile: .spur/run/${vars.__runId}-feature-dev-runall-answer.txt
           # Declared Layer-1 role (0538 R2): routing reason beside the agent: pin.
           role: planner
           timeoutMs: ${vars.stepTimeoutMs}
@@ -162,6 +140,8 @@ states:
         options:
           agent: ${vars.agent}
           input: /sp:dev-runall --tasks ${vars.featureTaskIds}
+          answerFile: .spur/run/${vars.__runId}-feature-dev-runall-answer.txt
+          expectFile: .spur/run/${vars.__runId}-feature-dev-runall-answer.txt
           # Declared Layer-1 role (0538 R2): routing reason beside the agent: pin.
           role: planner
           timeoutMs: ${vars.stepTimeoutMs}
@@ -177,17 +157,10 @@ states:
     onEnter:
       - kind: shell
         options:
+          # (e) 0825: 9 lines — `feature check --as done` runs EXACTLY ONCE per run (0782 R3)
+          # and owns the status write the guards read; split would re-run the check.
           command: >-
-            mkdir -p .spur/run;
-            VERIFY_JSON=".spur/run/$__runId-feature-dev-verify.json";
-            VERIFY_STATUS=".spur/run/$__runId-feature-dev-verify.status";
-            VERIFY_TMP=".spur/run/$__runId-feature-dev-verify.status.tmp";
-            rm -f "$VERIFY_STATUS";
-            set +e;
-            $spurBin feature check "$featureId" --as done --json > "$VERIFY_JSON" 2>&1;
-            check_rc=$?;
-            set -e;
-            if [ "$check_rc" -eq 0 ] && jq -e 'type == "array" and length > 0 and all(.[]; .pass == true)' "$VERIFY_JSON" > /dev/null 2>&1; then printf 'PASS\n' > "$VERIFY_TMP" && mv -f "$VERIFY_TMP" "$VERIFY_STATUS"; else printf 'FAIL\n' > "$VERIFY_TMP" && mv -f "$VERIFY_TMP" "$VERIFY_STATUS"; fi
+            mkdir -p .spur/run; rm -f ".spur/run/$__runId-feature-dev-verify.status"; $spurBin feature check "$featureId" --as done --json > ".spur/run/$__runId-feature-dev-verify.json" 2>&1 && jq -e 'type == "array" and length > 0 and all(.[]; .pass == true)' ".spur/run/$__runId-feature-dev-verify.json" > /dev/null 2>&1 && R=PASS || R=FAIL; printf '%s\n' "$R" > ".spur/run/$__runId-feature-dev-verify.status.tmp" && mv -f ".spur/run/$__runId-feature-dev-verify.status.tmp" ".spur/run/$__runId-feature-dev-verify.status"
       - kind: note
         options:
           message: "Feature verification for ${vars.featureId}: decision at .spur/run/${vars.__runId}-feature-dev-verify.status (evidence at .spur/run/${vars.__runId}-feature-dev-verify.json) — exactly one `feature check --as done --json` invocation per run (0782 R3)."
@@ -211,36 +184,18 @@ states:
       `requireCleanReview=true` turns a non-clean COLLECTED verdict (including
       collect FAIL) into the blocking edge declared first below.
     onEnter:
+      # (e) 0825: one request action + one collect action (same verdicts, same files,
+      # collect gated on the captured head SHA as before) to hold the §1.2 budgets.
       - kind: shell
         options:
           command: >-
-            mkdir -p .spur/run &&
-            STATUS_FILE=".spur/run/$__runId-integration-review.status" &&
-            REQUEST_JSON=".spur/run/$__runId-integration-review.json" &&
-            COLLECT_JSON=".spur/run/$__runId-integration-review-collect.json" &&
-            COLLECT_STATUS=".spur/run/$__runId-integration-review-collect.status" &&
-            PR_REQUEST_STATUS=".spur/run/$__runId-integration-review-pr-request.status" &&
-            set +e &&
-            bun "$(superskill script path sp pr-reviewing.ts)" request --base "$baseBranch" --json
-            --status-file "$PR_REQUEST_STATUS" > "$REQUEST_JSON" 2>&1;
-            request_rc=$?; set -e &&
-            if [ "$request_rc" -eq 0 ]; then
-              printf 'PASS\n' > "$STATUS_FILE";
-            else
-              printf 'FAIL\n' > "$STATUS_FILE";
-            fi &&
-            HEAD_SHA=$(jq -r '.head // empty' "$REQUEST_JSON" 2>/dev/null || true) &&
-            if [ -z "$HEAD_SHA" ]; then
-              echo "integration-review: request result carried no head SHA — collect not run, recorded FAIL" >&2;
-              printf 'FAIL\n' > "$COLLECT_STATUS";
-            else
-              set +e &&
-              bun "$(superskill script path sp pr-reviewing.ts)" collect --head "$HEAD_SHA" --json
-              --status-file "$COLLECT_STATUS" > "$COLLECT_JSON" 2>&1;
-              collect_rc=$?; set -e &&
-              echo "integration-review: collect rc=$collect_rc status=$(cat "$COLLECT_STATUS" 2>/dev/null || echo missing) — verdict captured to $COLLECT_JSON" >&2;
-            fi;
-            exit 0
+            mkdir -p .spur/run; bun "$(superskill script path sp pr-reviewing.ts)" request --base "$baseBranch" --json --status-file ".spur/run/$__runId-integration-review-pr-request.status" > ".spur/run/$__runId-integration-review.json" 2>&1 && printf 'PASS\n' > ".spur/run/$__runId-integration-review.status" || printf 'FAIL\n' > ".spur/run/$__runId-integration-review.status"; exit 0
+      - kind: shell
+        options:
+          # (e) 0825: 8 lines — collect stays head-locked to the request record and records
+          # its verdict status file; the rc/status echo is the audible soft-fail contract.
+          command: >-
+            if H=$(jq -er '.head // empty | select(. != "")' ".spur/run/$__runId-integration-review.json" 2>/dev/null); then bun "$(superskill script path sp pr-reviewing.ts)" collect --head "$H" --json --status-file ".spur/run/$__runId-integration-review-collect.status" > ".spur/run/$__runId-integration-review-collect.json" 2>&1; rc=$?; echo "integration-review: collect rc=$rc status=$(cat ".spur/run/$__runId-integration-review-collect.status" 2>/dev/null || echo missing) — verdict captured to .spur/run/$__runId-integration-review-collect.json" >&2; else echo "integration-review: request result carried no head SHA — collect not run, recorded FAIL" >&2; printf 'FAIL\n' > ".spur/run/$__runId-integration-review-collect.status"; fi; exit 0
       - kind: note
         options:
           message: "Integration review for feature ${vars.featureId}: request verdict at .spur/run/${vars.__runId}-integration-review.status (request state only — REQUESTED/ALREADY_* are NEVER clean evidence); collected verdict at .spur/run/${vars.__runId}-integration-review-collect.status (CLEAN/FINDINGS/PENDING — the only clean evidence). Advisory unless requireCleanReview=true."

package/config/workflows/history-anatomy.yaml

@@ -72,7 +72,10 @@ vars:
   agent: "claude"
   spurBin: "spur"
   __runId: ""
-  stepTimeoutMs: "1800000"
+  # 2026-09-12: raised 30m -> 60m. Observed runs on zai/glm-5.3-flash butt against the
+  # 30m ceiling (correct exited at exactly 30m00s; a prior run passed at 29m52s) — the
+  # enrich/validate/correct stages legitimately need >30m for the 60KB candidate report.
+  stepTimeoutMs: "3600000"
   workflowFile: "config/workflows/history-anatomy.yaml"
   contractVersion: "1"
   reportDir: "docs/report"
@@ -225,6 +228,24 @@ states:
           role: reviewer
           expectFile: .spur/run/${vars.__runId}-validation.txt
           timeoutMs: ${vars.stepTimeoutMs}
+      - kind: shell
+        options:
+          # 2026-09-13 verdict-placement normalization: the 0771 guard reads ONLY the
+          # final line, but models sometimes lead with `Verdict: PASS` and close with
+          # prose — a substantive PASS must not die as a format phantom (observed
+          # 2026-09-13 run 94c4d6dc). Moves the verdict line to the end ONLY when
+          # exactly one exact-line verdict exists and it is PASS. Any `Verdict: FAIL`
+          # line anywhere (or multiple/ambiguous verdict lines) suppresses
+          # normalization, so the guard still fails and a real FAIL can never be
+          # laundered into a PASS.
+          command: >-
+            f=.spur/run/$__runId-validation.txt;
+            if [ -f "$f" ] && ! tail -n 1 "$f" | grep -qx 'Verdict: PASS'; then
+              p=$(grep -cx 'Verdict: PASS' "$f"); q=$(grep -cx 'Verdict: FAIL' "$f");
+              if [ "$p" = 1 ] && [ "$q" = 0 ]; then
+                grep -vx 'Verdict: PASS' "$f" > "$f.tmp" && printf 'Verdict: PASS\n' >> "$f.tmp" && mv "$f.tmp" "$f";
+              fi;
+            fi
       - kind: shell
         options:
           command: >-

package/config/workflows/idea-pipeline.yaml

@@ -331,7 +331,7 @@ states:
       - kind: agent.run
         options:
           agent: ${vars.planningAgent}
-          input: "Run sp:spec-decomposition for feature ${vars.featureId}. Read the brainstorm artifact, feature AC, and design doc. SIZING FIRST, before any JSON: apply the skill's `Default to NOT decomposing` rubric to the whole unit of work — if it scores 0-2 the correct output is a ONE-entry batch, not many. Scenario count is not task count: merge scenarios that one task delivers (same file surface, same subsystem, or unreadable apart in review), and list every scenario a task covers in its background. Merging never costs AC coverage — one task may carry several scenarios. Do not emit one entry per scenario or per requirement by reflex. Then produce a task-batch JSON array at .spur/run/${vars.__runId}-idea-task-batch.json, validated against task-batch.schema.json. Schema-permitted fields per entry: `name`, `background`, `requirements`, `design`, `plan`, `acceptance_criteria`, `feature_id`, `parent_wbs`, `priority`, `tags`, `template` — schema validation rejects anything else. `design`, `plan`, and `acceptance_criteria` are supported batch fields and normal default planning fills them from your analysis; the per-task refine step after batch-create still deepens them when a task needs more detail. Validate locally against the schema before emitting. Also emit the private task-order sidecar at .spur/run/${vars.__runId}-idea-task-order.json: a JSON array (one entry per batch item) of `{ name: <exact batch item name>, depends_on_names: [<batch item names>] }` declaring ordering/dependencies between the batch items; use `[]` when no ordering exists. Every `name` and every dependency must match exactly one batch item `name` — it is private workflow data, not part of task-batch.schema.json."
+          input: "Run sp:spec-decomposition for feature ${vars.featureId} per skill references/decomposition.md § Idea-pipeline emission: sizing first, then the batch JSON at .spur/run/${vars.__runId}-idea-task-batch.json and the private task-order sidecar at .spur/run/${vars.__runId}-idea-task-order.json."
           # Declared Layer-1 role (0538 R2): routing reason beside the agent: pin.
           role: planner
           expectFile: .spur/run/${vars.__runId}-idea-task-batch.json
@@ -341,6 +341,8 @@ states:
       # unique, every sidecar name/dependency must refer to exactly one batch name, and — the
       # converse (F2, 0518 verify) — every batch name must appear in the sidecar, so a partial
       # sidecar can never silently skip `task deps` for an unlisted item (`[]` is valid).
+      # (warn) stays inline by design: one jq predicate over two run-scoped files; splitting
+      # it into a script would hide the exact fail-closed contract this guard pins (0824).
       - kind: shell
         options:
           command: >-
@@ -380,17 +382,22 @@ states:
       (CLI error or malformed result) writes the failed sentinel, preserving the
       existing retry behavior.
     onEnter:
+      # (e) idempotent batch creation: sentinel + temp/mv atomicity stay inline (0824);
+      # the jq verdict check keeps the done sentinel truthful so handoff-finalize can
+      # zip batch names to the returned WBS list.
       - kind: shell
         options:
           command: >-
-            if test -f .spur/run/$__runId-idea-batch-create.done; then exit 0; fi;
-            rm -f .spur/run/$__runId-idea-batch-create.failed .spur/run/$__runId-idea-batch-create-result.json .spur/run/$__runId-idea-batch-create-result.json.tmp;
-            if $spurBin task batch-create --file .spur/run/$__runId-idea-task-batch.json --skip-ready --json > .spur/run/$__runId-idea-batch-create-result.json.tmp && jq -e ".created == (.wbs | length)" .spur/run/$__runId-idea-batch-create-result.json.tmp >/dev/null 2>&1; then
-              mv .spur/run/$__runId-idea-batch-create-result.json.tmp .spur/run/$__runId-idea-batch-create-result.json &&
-              date -u +%Y-%m-%dT%H:%M:%SZ > .spur/run/$__runId-idea-batch-create.done;
+            P=".spur/run/$__runId-idea-batch-create" &&
+            test ! -f "$P.done" || exit 0 &&
+            rm -f "$P.failed" "$P-result.json" "$P-result.json.tmp" &&
+            if $spurBin task batch-create --file .spur/run/$__runId-idea-task-batch.json --skip-ready --json > "$P-result.json.tmp" &&
+              jq -e ".created == (.wbs | length)" "$P-result.json.tmp" >/dev/null 2>&1; then
+              mv "$P-result.json.tmp" "$P-result.json" &&
+              date -u +%Y-%m-%dT%H:%M:%SZ > "$P.done";
             else
-              rm -f .spur/run/$__runId-idea-batch-create-result.json.tmp;
-              date -u +%Y-%m-%dT%H:%M:%SZ > .spur/run/$__runId-idea-batch-create.failed;
+              rm -f "$P-result.json.tmp" &&
+              date -u +%Y-%m-%dT%H:%M:%SZ > "$P.failed";
             fi
 
   - id: ready-prepare
@@ -411,24 +418,22 @@ states:
           # Declared Layer-1 role (0538 R2), same planner executor as decompose.
           role: planner
           timeoutMs: ${vars.stepTimeoutMs}
+          # (warn) non-slash pointer: the 0788 checklist is per-checkout (digest via the
+          # project's own computePlanningDigest), so the bounded prompt pins the skill
+          # reference instead of a command; artifacts are gated by answerFile/expectFile.
           input: >-
-            Ready-by-default preparation for feature ${vars.featureId} (0788). The created WBS
-            values are in .spur/run/${vars.__runId}-idea-batch-create-result.json under .wbs.
-            For EACH wbs: resolve the task file with `spur task path <wbs> --json`; open the
-            task doc and apply the ready-refinement checklist — make requirements, design,
-            plan, acceptance criteria, decisions, dependencies and premises present and
-            non-placeholder so `spur task check <wbs> --json` exits 0; edit only planning
-            sections, never Solution/Testing/Review/History. Record one checklist row per id
-            with concrete evidence of how you verified it. Compute the planning digest with
-            the project's own implementation when this is a monorepo checkout:
-            `bun -e 'const m = await import("./packages/app/src/services/task-readiness"); console.log(m.computePlanningDigest(await Bun.file(process.argv[1]).text()))' <task-file>`
-            — when that is impossible in this checkout, set status "skipped" instead of
-            guessing a digest. Finally write .spur/run/${vars.__runId}-idea-ready.json with
-            exactly this shape: {"runId":"${vars.__runId}","depth":"ready","tasks":[{"wbs":"<wbs>","status":"ready" or "failed" or "skipped","planningDigest":"<sha256 hex>","checks":[{"id":"requirements" or "design" or "plan" or "ac" or "decisions" or "dependencies" or "premises","pass":true or false,"evidence":"<how verified>"}]}]}. A task you cannot fully prepare gets status "failed" or "skipped" — never fabricate evidence; the handoff degrades to refineall.
+            Run the ready-prepare stage for feature ${vars.featureId} per sp:spur-dev
+            references/planning-workflow.md § Step 5.6 (Ready preparation): read
+            .spur/run/${vars.__runId}-idea-batch-create-result.json and write
+            .spur/run/${vars.__runId}-idea-ready.json.
+          answerFile: .spur/run/${vars.__runId}-ready-prepare-answer.txt
+          expectFile: .spur/run/${vars.__runId}-ready-prepare-answer.txt
       # Fail-closed shape validation (mirrors the order-sidecar guard). Absence is
       # normalized to an empty sidecar: finalize and the seeded fallback then degrade
       # the recommendation to refineall instead of failing the run. A PRESENT but
       # malformed sidecar fails the run here — it must not masquerade as evidence.
+      # (warn) stays inline by design: single jq shape predicate over the run-scoped
+      # sidecar; the failed/skipped rows are what degrade the recommendation (0824).
       - kind: shell
         options:
           command: >-
@@ -447,87 +452,23 @@ states:
       A mapping or CLI error fails the run before terminal handoff; an unready
       task is a successful planning outcome recorded as a refineall recommendation.
     onEnter:
-      # F1 (0518 verify): the per-task check loop fails closed. A stderr-only (non-JSON)
-      # `task check --json` exception would otherwise drop that task from the JSONL results
-      # without failing the run (the loop's exit status is its last body command), silently
-      # flipping the recommendation to runall — violating R3's any-fail=>refineall invariant.
-      # Each `jq -c … >> "$CHECKS"` appends under `|| exit 1`, and a row-count assertion
-      # (CHECKS lines == WBS count) gates the recommendation computation.
-      # D5-O (R4): the deterministic writer owns finalization wherever it exists.
-      # `idea-handoff-cli.ts` calls the tested `finalizeIdeaHandoff` capability and
-      # implements this exact contract (equal-length/unique-name zip, `spur task deps`,
-      # `spur feature refresh`, per-task check with a row-count assertion, and a single
-      # recommended next command). `spur init` never scaffolds `packages/`, so seeded
-      # projects fall through to the portable shell program below — the same
-      # monorepo-writer/shell-fallback split the wrap-up metrics hop uses (0604 Q&A).
+      # (d) idea-handoff owns finalization (0824): the bundled finalizeIdeaHandoff
+      # capability runs from the monorepo checkout first; seeded projects (no
+      # packages/, no plugins/) resolve the registered idea-handoff.mjs twin under
+      # bare node; neither present fails closed (exit 1) instead of silently
+      # skipping finalization. The zip/deps/refresh/check/report contract itself is
+      # pinned by finalizeIdeaHandoff unit tests, not by this file.
       - kind: shell
         options:
           command: >-
             if [ -f packages/app/src/workflow/idea-handoff-cli.ts ]; then
               bun packages/app/src/workflow/idea-handoff-cli.ts;
-              exit $?;
-            fi;
-            BATCH=".spur/run/$__runId-idea-task-batch.json" &&
-            RESULT=".spur/run/$__runId-idea-batch-create-result.json" &&
-            ORDER=".spur/run/$__runId-idea-task-order.json" &&
-            REPORT=".spur/run/$__runId-idea-handoff.md" &&
-            DEPMAP=".spur/run/$__runId-idea-dep-map.tsv" &&
-            CHECKS=".spur/run/$__runId-idea-check-results.jsonl" &&
-            rm -f "$DEPMAP" "$CHECKS" &&
-            jq -e --slurpfile b "$BATCH" '(.wbs | length) == ($b[0] | length) and (($b[0] | map(.name) | unique | length) == ($b[0] | length))' "$RESULT" >/dev/null &&
-            jq -r --slurpfile b "$BATCH" --slurpfile r "$RESULT" --slurpfile o "$ORDER" '
-              ($b[0] | map(.name)) as $names | ($r[0].wbs) as $wbs |
-              $o[0] | .[] |
-              (.name as $n | $names | index($n)) as $i |
-              (if $i == null or $wbs[$i] == null then "MISSING" else $wbs[$i] end) as $own |
-              ((.depends_on_names // []) | map(. as $dep | $names | index($dep) as $d | if $d == null then "MISSING" else $wbs[$d] end) | join(" ")) as $deps |
-              [.name, $own, $deps] | @tsv
-            ' "$ORDER" > "$DEPMAP" &&
-            while IFS="$(printf '\t')" read -r name own deps; do
-              test "$own" != MISSING || exit 1;
-              if test -n "$deps"; then
-                for d in $deps; do test "$d" != MISSING || exit 1; done;
-                $spurBin task deps "$own" set $deps --json >/dev/null || exit 1;
-              fi;
-            done < "$DEPMAP" &&
-            $spurBin feature refresh --feature "$featureId" --json >/dev/null &&
-            WBS_LIST=$(jq -r '.wbs | join(" ")' "$RESULT") &&
-            for wbs in $WBS_LIST; do
-              TMP=".spur/run/$__runId-idea-check-$wbs.tmp";
-              if $spurBin task check "$wbs" --json > "$TMP" 2>&1; then P=true; else P=false; fi;
-              jq -c --arg w "$wbs" --argjson p "$P" 'if (type == "array" and length > 0) then {wbs: $w, pass: $p, status: (.[0].status // "unknown")} else {wbs: $w, pass: $p, status: "missing"} end' "$TMP" >> "$CHECKS" || exit 1;
-              rm -f "$TMP";
-            done &&
-            CHECK_ROWS=$(wc -l < "$CHECKS" | tr -d ' ') &&
-            WBS_COUNT=$(printf '%s\n' $WBS_LIST | wc -l | tr -d ' ') &&
-            test "$CHECK_ROWS" = "$WBS_COUNT" &&
-            READY=".spur/run/$__runId-idea-ready.json" &&
-            (test -f "$READY" || printf '{"tasks":[]}\n' > "$READY") &&
-            NEXT=$(jq -r --arg feature "$featureId" --slurpfile c "$CHECKS" --slurpfile k "$READY" 'if (($c | any(.[]; .pass == false)) or (($k[0].tasks // []) | length == 0) or (($k[0].tasks // []) | any(.[]; .status != "ready"))) then "/sp:dev-refineall --feature \($feature) --auto --depth ready" else "/sp:dev-runall --feature \($feature) --auto" end' "$RESULT") &&
-            {
-              echo "# Idea pipeline handoff report";
-              echo;
-              echo "Feature: $featureId";
-              echo "Run ID: $__runId";
-              echo;
-              echo "## Created tasks";
-              printf '%s\n' $WBS_LIST | sed 's/^/  - /';
-              echo;
-              echo "## Per-task readiness (spur task check)";
-              echo;
-              echo "| WBS | Outcome |";
-              echo "|-----|---------|";
-              while IFS= read -r line; do
-                w=$(printf '%s' "$line" | jq -r '.wbs');
-                o=$(printf '%s' "$line" | jq -r 'if .pass then "PASS" else "FAIL" end');
-                printf '| %s | %s |\n' "$w" "$o";
-              done < "$CHECKS";
-              echo;
-              echo "## Next command";
-              echo;
-              echo "$NEXT";
-            } > "$REPORT"
-
+            elif H="$(superskill script path sp idea-handoff.mjs 2>/dev/null)" && [ -f "$H" ]; then
+              node "$H";
+            else
+              echo "idea handoff failed closed — idea-handoff script not found — run 'superskill install sp'" >&2;
+              exit 1;
+            fi
   - id: handoff
     description: >
       Terminal — idea pipeline complete. Ordering applied, roster refreshed, and
@@ -627,6 +568,8 @@ transitions:
   - from: ac-generate
     to: system-design
     description: "profile=auto, check passed, design route — run system design."
+    # (warn) 4 test segments: profile gate + captured status + route signal jointly own the
+    # route; guards never re-run the CLI (0769), so there is nothing smaller to extract.
     guard:
       kind: shell
       options:
@@ -635,6 +578,8 @@ transitions:
   - from: ac-generate
     to: decompose
     description: "profile=auto, check passed, skip-design route — go directly to decompose."
+    # (warn) 5 test segments: same as auto-skip 1 plus the OR'd design=auto/needs_design=false
+    # pair that keeps one captured signal file authoritative for both routes (0769).
     guard:
       kind: shell
       options:
@@ -643,6 +588,8 @@ transitions:
   - from: ac-generate
     to: ac-generate
     description: "profile=auto, check failed, retry cap not reached — re-run ac-generate."
+    # (warn) 4 test segments: profile gate + captured status + captured retry count; the
+    # retry loop is the smallest honest formulation of cap<3 routing (0769).
     guard:
       kind: shell
       options:
@@ -651,6 +598,8 @@ transitions:
   - from: ac-generate
     to: failed
     description: "profile=auto, check failed after 3 retries — escalate to failed."
+    # (warn) 4 test segments: mirror of the retry guard with cap>=3; keeping escalation and
+    # retry as one test-chain pair makes the cap boundary auditable in the diff (0769).
     guard:
       kind: shell
       options:
@@ -667,6 +616,8 @@ transitions:
   - from: feature-check
     to: system-design
     description: Feature check passed, design route — run system design.
+    # (warn) 4 test segments: HITL answer + captured status + design route; same shape as the
+    # ac-generate auto-skips, reading the same captured signal files (0769).
     guard:
       kind: shell
       options:
@@ -674,6 +625,8 @@ transitions:
   - from: feature-check
     to: decompose
     description: Feature check passed, skip-design route — go directly to decompose.
+    # (warn) 5 test segments: same as the feature-check design route plus the OR'd
+    # design=auto/needs_design=false pair (0769).
     guard:
       kind: shell
       options:
@@ -681,6 +634,8 @@ transitions:
   - from: feature-check
     to: ac-generate
     description: "Feature check failed — revise AC (retry cap: 3)."
+    # (warn) 4 test segments: HITL reject/failed-status OR-pair + captured retry count; the
+    # interactive retry mirror of the ac-generate cap<3 guard (0769).
     guard:
       kind: shell
       options:
@@ -688,6 +643,7 @@ transitions:
   - from: feature-check
     to: failed
     description: Feature check failed after 3 retries — escalate to failed.
+    # (warn) 4 test segments: mirror of the revise guard with cap>=3 (0769).
     guard:
       kind: shell
       options: