@gobing-ai/spur 0.3.34 → 0.3.36

package/config/corpus-baseline.json

@@ -0,0 +1,61 @@
+{
+    "note": "Accepted `spur task check` / `spur feature check` errors. This is a POLICY file, not a status dump: it is an INPUT to `bun run corpus-check`, so it must be committed — an untracked baseline means the gate behaves differently on every clone and CI, and the first person to meet 8 unexplained failures disables it. Do not move it under `.spur/` (gitignored, ephemeral run artifacts). The gate fails on any error NOT listed here, and ALSO on any entry that no longer reproduces, so this cannot rot into a silent suppression list. Every entry needs a real diagnosis, a reason, and a date — 'pre-existing, not investigated' is not a reason.",
+    "entries": [
+        {
+            "kind": "task",
+            "id": "0368",
+            "code": "L3.required-section-placeholder",
+            "reason": "RATCHET DRIFT, not a bypass. Closed 2026-07-28; the rule that flags it landed 2026-08-01 (f373e90b), four days later. It was legally closed under the rules of its day and became non-compliant retroactively. Backfill Testing/Solution or exempt permanently when someone next touches it.",
+            "since": "2026-08-07"
+        },
+        {
+            "kind": "task",
+            "id": "0454",
+            "code": "L3.solution-file-line",
+            "reason": "GENUINE PRE-FIX BYPASS. Closed 2026-08-06, well after the rule existed (2026-07-24), via the `--no-lifecycle` hole closed on 2026-08-07. Its Solution needs one file:line citation. The only entry here representing real unpaid debt.",
+            "since": "2026-08-07"
+        },
+        {
+            "kind": "feature",
+            "id": "M",
+            "code": "L3.ac-bdd-error",
+            "reason": "CHECKER GAP, not corpus debt. M is a wayfinder MAP; its Acceptance Criteria section carries an explicit prose disclaimer ('This feature is a wayfinder MAP... It has no acceptance criteria — its target is the ## Goal'). The BDD validator requires a `Feature:` declaration and has no notion of the map feature type. Correct fix is to teach `feature check` about wayfinder maps, not to bolt fake Gherkin onto a map.",
+            "since": "2026-08-07"
+        },
+        {
+            "kind": "feature",
+            "id": "M",
+            "code": "L3.ac-bdd-invalid",
+            "reason": "Downstream of the M ac-bdd-error entry — same missing-Feature-declaration cause, reported twice by the validator.",
+            "since": "2026-08-07"
+        },
+        {
+            "kind": "feature",
+            "id": "F82",
+            "code": "L3.ac-bdd-error",
+            "reason": "CHECKER GAP, same class as M: wayfinder map with a deliberate no-AC prose disclaimer (plus a noted exception for two implementation-ready child tickets). Not a syntax error to fix.",
+            "since": "2026-08-07"
+        },
+        {
+            "kind": "feature",
+            "id": "F82",
+            "code": "L3.ac-bdd-invalid",
+            "reason": "Downstream of the F82 ac-bdd-error entry — same cause, reported twice.",
+            "since": "2026-08-07"
+        },
+        {
+            "kind": "feature",
+            "id": "F821",
+            "code": "L3.ac-bdd-error",
+            "reason": "LEGACY AC FORMAT, not a map. F821's Acceptance Criteria are written as plain `R1 @core — …` prose rows rather than Gherkin, an older convention predating the BDD validator. Converting them is a real (small) job with traceability risk: AC titles are identity keys, so renaming breaks child-task scenario references.",
+            "since": "2026-08-07"
+        },
+        {
+            "kind": "feature",
+            "id": "F821",
+            "code": "L3.ac-bdd-invalid",
+            "reason": "Downstream of the F821 ac-bdd-error entry — same cause, reported twice.",
+            "since": "2026-08-07"
+        }
+    ]
+}

package/config/launchd/ai.gobing.spur.history.daily.plist

@@ -0,0 +1,56 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
+<plist version="1.0">
+<dict>
+    <key>Label</key>
+    <string>ai.gobing.spur.history.daily</string>
+
+    <!-- ── Program: run the nightly history pipeline (task 0470+) ──────────── -->
+    <!-- Replace SPUR_BIN with the absolute path to your `spur` binary          -->
+    <!-- (e.g. /opt/homebrew/bin/spur or the bun-linked monorepo CLI).           -->
+    <!-- Replace PROJECT_DIR with the absolute path to your Spur project root.   -->
+    <key>ProgramArguments</key>
+    <array>
+        <string>SPUR_BIN</string>
+        <string>history</string>
+        <string>daily</string>
+        <string>--json</string>
+    </array>
+
+    <!-- Working directory: the Spur project root. The daily command resolves    -->
+    <!-- .spur/reports, .spur/logs, and the SQLite ledger relative to this.       -->
+    <key>WorkingDirectory</key>
+    <string>PROJECT_DIR</string>
+
+    <!-- ── Schedule: daily wall-clock trigger (R3) ─────────────────────────── -->
+    <!-- 02:00 local time — picks a low-activity hour. Adjust Hour to taste.      -->
+    <!-- launchd survives reboot and runs missed CalendarInterval jobs at the     -->
+    <!-- next login, so a laptop closed overnight self-heals.                     -->
+    <key>StartCalendarInterval</key>
+    <dict>
+        <key>Hour</key>
+        <integer>2</integer>
+        <key>Minute</key>
+        <integer>0</integer>
+    </dict>
+
+    <!-- ── Log capture (R4): failures before Spur's own logging initializes ── -->
+    <!-- are still written here. launchd owns these file handles from the first   -->
+    <!-- byte — there is no window where output is lost.                          -->
+    <key>StandardOutPath</key>
+    <string>PROJECT_DIR/.spur/logs/history-daily.out</string>
+    <key>StandardErrorPath</key>
+    <string>PROJECT_DIR/.spur/logs/history-daily.err</string>
+
+    <!-- RunAtLoad=false: do NOT fire every time the agent loads (boot/login).   -->
+    <!-- StartCalendarInterval already re-fires a missed slot when the machine   -->
+    <!-- wakes, so RunAtLoad would only cause duplicate runs.                    -->
+
+    <!-- If the process exits non-zero, do NOT auto-restart — the daily command  -->
+    <!-- has its own retry semantics (checkpoint resume), and a tight retry loop  -->
+    <!-- would mask a configuration error. The operator checks the four detection -->
+    <!-- layers (R5) instead.                                                     -->
+    <key>KeepAlive</key>
+    <false/>
+</dict>
+</plist>

package/config/rules/strict/runtime-boundaries.yaml

@@ -60,6 +60,7 @@ rules:
             - "packages/app/src/services/token-ledger-service.ts" # FD byte-window log tailing
             - "packages/app/src/services/token-ledger-watcher.ts" # node:fs watch() live watcher
             - "packages/app/src/services/project-registry.ts" # atomic projects.json persistence
+            - "packages/app/src/services/history-service.ts" # versioned analyze artifact + bounded-errors sidecar + latest.json symlink pointer (task 0474); ts-runtime FileSystem seam has no symlink, so the pointer uses node:fs directly (mirrors project-registry.ts persistence exemption)
             - "packages/app/src/observability/workflow-run-log-sink.ts" # sync FD append for mid-run tail-able all-in-one run log (task 0426 / feature D2); append() is sync from the observability bus
             - "apps/cli/src/commands/workflow.ts" # FD byte-window tail of the mid-run run log for `workflow trace --follow` streaming (task 0428 / feature D2); readSync at offset over the observability sink's FDs
             - "apps/web/src/modules/discover.ts" # Vite/Astro module scanner fallback under bun test

package/config/templates/task/brainstorm.md

@@ -10,6 +10,7 @@ parent_wbs: null
 priority: P2
 tags: []
 dependencies: []
+ac_numbering: task-local
 created_at: "{{ CREATED_AT }}"
 updated_at: "{{ CREATED_AT }}"
 ---

package/config/templates/task/feature-impl.md

@@ -10,6 +10,7 @@ parent_wbs: null
 priority: P2
 tags: []
 dependencies: []
+ac_numbering: task-local
 created_at: "{{ CREATED_AT }}"
 updated_at: "{{ CREATED_AT }}"
 ---
@@ -30,7 +31,9 @@ updated_at: "{{ CREATED_AT }}"
 
 ### Q&A
 
-<!-- Clarifications and decisions made during refinement. Keep empty if none. -->
+<!-- CLOSED decisions from refinement: what was chosen and why, what was deferred and on what
+     condition. Not a parking lot for open questions — an unanswered question here means the task
+     is not ready to hand off. Keep empty if none. -->
 
 ### Design
 

package/config/templates/task/issue.md

@@ -10,6 +10,7 @@ parent_wbs: null
 priority: P2
 tags: ["bug"]
 dependencies: []
+ac_numbering: task-local
 created_at: "{{ CREATED_AT }}"
 updated_at: "{{ CREATED_AT }}"
 ---

package/config/templates/task/meta.md

@@ -10,6 +10,7 @@ parent_wbs: null
 priority: P2
 tags: ["meta"]
 dependencies: []
+ac_numbering: task-local
 created_at: "{{ CREATED_AT }}"
 updated_at: "{{ CREATED_AT }}"
 ---
@@ -30,7 +31,9 @@ updated_at: "{{ CREATED_AT }}"
 
 ### Q&A
 
-<!-- Clarifications and decisions made during refinement. Keep empty if none. -->
+<!-- CLOSED decisions from refinement: what was chosen and why, what was deferred and on what
+     condition. Not a parking lot for open questions — an unanswered question here means the task
+     is not ready to hand off. Keep empty if none. -->
 
 ### Design
 

package/config/templates/task/review.md

@@ -10,6 +10,7 @@ parent_wbs: null
 priority: P2
 tags: ["review"]
 dependencies: []
+ac_numbering: task-local
 created_at: "{{ CREATED_AT }}"
 updated_at: "{{ CREATED_AT }}"
 ---

package/config/templates/task/standard.md

@@ -10,6 +10,7 @@ parent_wbs: null
 priority: P2
 tags: []
 dependencies: []
+ac_numbering: task-local
 created_at: "{{ CREATED_AT }}"
 updated_at: "{{ CREATED_AT }}"
 ---
@@ -30,7 +31,9 @@ updated_at: "{{ CREATED_AT }}"
 
 ### Q&A
 
-<!-- Clarifications and decisions made during refinement. Keep empty if none. -->
+<!-- CLOSED decisions from refinement: what was chosen and why, what was deferred and on what
+     condition. Not a parking lot for open questions — an unanswered question here means the task
+     is not ready to hand off. Keep empty if none. -->
 
 ### Design
 

package/config/workflows/idea-pipeline.yaml

@@ -247,7 +247,7 @@ states:
       - kind: agent.run
         options:
           agent: ${vars.agent}
-          input: "Run sp:spec-decomposition for feature ${vars.featureId}. Read the brainstorm artifact, feature AC, and design doc. Produce a task-batch JSON array at .spur/run/${vars.__runId}-idea-task-batch.json, validated against apps/cli/schemas/task-batch.schema.json. Schema-permitted fields per entry: `name`, `background`, `requirements`, `feature_id`, `parent_wbs`, `priority`, `tags`, `template` — schema validation rejects anything else. Acceptance Criteria, Design, and Plan sections are filled in by the per-task refine step after batch-create, NOT at decompose time. Validate locally against the schema before emitting."
+          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 apps/cli/schemas/task-batch.schema.json. Schema-permitted fields per entry: `name`, `background`, `requirements`, `feature_id`, `parent_wbs`, `priority`, `tags`, `template` — schema validation rejects anything else. Acceptance Criteria, Design, and Plan sections are filled in by the per-task refine step after batch-create, NOT at decompose time. Validate locally against the schema before emitting."
           expectFile: .spur/run/${vars.__runId}-idea-task-batch.json
           timeoutMs: ${vars.stepTimeoutMs}
 

package/config/workflows/task-pipeline.yaml

@@ -57,6 +57,10 @@ vars:
   # `--vars '{"agent":"claude"}'`. Pinned (not left to the AiRunner's <default>
   # selection) so a broken/misconfigured agent on the box can't silently capture the run.
   agent: "omp"
+  # Implement-only executor override (R1, task 0454). Defaults to same as agent.
+  # Override with `--vars '{"implementAgent":"omp-zai"}'` to pin implement to a
+  # specific executor while other hops keep the default agent.
+  implementAgent: "omp"
   # Step-level timeout for agentic hops (review / verify / test-fix) in ms.
   # Soft quality-gate shells are unbounded by this var (host shell only).
   # Raised 600s → 1800s (task 0398 R4 / H6 dogfood). Override:
@@ -78,7 +82,7 @@ vars:
   # and the fixall slash input all use this same var so the command stays single-sourced.
   # TRUSTED CONFIG ONLY — this string is executed via `sh -c` (see test/test-recheck). Never
   # interpolate untrusted operator/LLM input into qualityGateCmd (task 0436 SECUA residual).
-  qualityGateCmd: "bun run autofix && bun run spur-check"
+  qualityGateCmd: "bun run format && bun run spur-check"
   # Max /sp:dev-fixall attempts after a red quality-gate probe/recheck (bounded; no thrash).
   # Attempt counter: .spur/run/<wbs>-test-fix-attempt. Default 2 = two fixall hops before failed.
   qualityGateMaxFixAttempts: "2"
@@ -86,6 +90,12 @@ vars:
   # project can point it at its own formatter; invoked best-effort (a missing or
   # failing formatter must never abort a run — the quality gate is the real gate).
   formatCmd: "bun run format"
+  # Max R-items in ## Requirements before size precheck fails (R2, task 0454).
+  # Override with `--vars '{"maxImplementReqs":"12"}'`.
+  maxImplementReqs: "5"
+  # Max checklist items under ## Plan before size precheck fails (R2, task 0454).
+  # Override with `--vars '{"maxImplementPlanItems":"15"}'`.
+  maxImplementPlanItems: "8"
 
 states:
   - id: precheck
@@ -128,6 +138,17 @@ states:
               fi;
             fi;
             exit 0
+      # R2 (0454): task size precheck — evaluate R-item and Plan-item counts.
+      # Writes PASS/FAIL to .spur/run/<wbs>-precheck-size.status. Always exit 0
+      # (soft check, like doctor). The precheck→implement guard reads the file.
+      - kind: shell
+        options:
+          command: >-
+            SIZE_FILE=".spur/run/$wbs-precheck-size.status" &&
+            mkdir -p .spur/run &&
+            bun plugins/sp/scripts/task-size-precheck.ts "$wbs"
+            --max-reqs "$maxImplementReqs" --max-plan-items "$maxImplementPlanItems" &&
+            exit 0
 
   - id: implement
     description: >
@@ -145,7 +166,7 @@ states:
     onEnter:
       - kind: agent.run
         options:
-          agent: ${vars.agent}
+          agent: ${vars.implementAgent}
           # Pure slash command only (ADR-043). Anti-recursion / implement discipline
           # lives in /sp:dev-run --mode implement → sp:code-implementation, not YAML prose.
           input: /sp:dev-run --mode implement ${vars.wbs} --auto
@@ -269,8 +290,10 @@ states:
   - id: verify
     description: >
       Functional verification (BDD + traceability) via /sp:dev-verify. The agent's
-      captured answer is written to `.spur/run/<wbs>-verify-answer.txt`; a deterministic
-      shell step then derives the verdict and writes the gate artifact
+      captured answer is written to `.spur/run/<wbs>-verify-answer.txt` and MUST follow
+      the answer-file schema contract (explicit `Verdict: PASS|PARTIAL|FAIL` line plus
+      `| Req | Status | Evidence |` and `| AC | Status | Evidence Type | Evidence |` tables);
+      a deterministic shell step then derives the verdict and writes the gate artifact
       `.spur/run/<wbs>-verdict.json` (so the completion gate reads a real verdict, not
       agent discretion — R9). The verdict is PASS only if the agent both reported PASS
       AND `spur task check` passes; otherwise FAIL.
@@ -352,7 +375,7 @@ transitions:
     guard:
       kind: shell
       options:
-        command: 'test "$(cat .spur/run/$wbs-precheck-doctor.status 2>/dev/null)" = PASS && $spurBin task check $wbs'
+        command: 'test "$(cat .spur/run/$wbs-precheck-doctor.status 2>/dev/null)" = PASS && test "$(cat .spur/run/$wbs-precheck-size.status 2>/dev/null)" = PASS && $spurBin task check $wbs'
   - from: precheck
     to: failed
     description: Doctor FAIL and/or task check failed — stop before implement.

package/package.json

@@ -1,7 +1,7 @@
 {
     "name": "@gobing-ai/spur",
-    "version": "0.3.34",
-    "description": "Spur CLI \u2014 local-first harness for mainstream coding agents: constraint checking, workflow orchestration, agent health, and history analytics. Bun-native; exposes the `spur` command.",
+    "version": "0.3.36",
+    "description": "Spur CLI — local-first harness for mainstream coding agents: constraint checking, workflow orchestration, agent health, and history analytics. Bun-native; exposes the `spur` command.",
     "keywords": [
         "spur",
         "cli",
@@ -50,14 +50,14 @@
     },
     "devDependencies": {
         "@commander-js/extra-typings": "^14.0.0",
-        "@gobing-ai/ts-db": "^0.4.19",
-        "@gobing-ai/ts-ai-runner": "^0.4.19",
-        "@gobing-ai/ts-dual-workflow-engine": "^0.4.19",
-        "@gobing-ai/ts-infra": "^0.4.19",
-        "@gobing-ai/ts-llm-jsonl-importer": "^0.4.19",
-        "@gobing-ai/ts-rule-engine": "^0.4.19",
-        "@gobing-ai/ts-runtime": "^0.4.19",
-        "@gobing-ai/ts-utils": "^0.4.19",
+        "@gobing-ai/ts-db": "^0.4.23",
+        "@gobing-ai/ts-ai-runner": "^0.4.23",
+        "@gobing-ai/ts-dual-workflow-engine": "^0.4.23",
+        "@gobing-ai/ts-infra": "^0.4.23",
+        "@gobing-ai/ts-llm-jsonl-importer": "^0.4.23",
+        "@gobing-ai/ts-rule-engine": "^0.4.23",
+        "@gobing-ai/ts-runtime": "^0.4.23",
+        "@gobing-ai/ts-utils": "^0.4.23",
         "@types/bun": "1.3.14",
         "@types/figlet": "^1.7.0",
         "@types/node-notifier": "8.0.5",