create-pmos 0.34.0 → 0.36.0

package/bin/create-pmos.js

@@ -129,15 +129,27 @@ function workspaceOrDie() {
   }
 }
 
+// A spawn that never RAN is a failure, not a success: spawnSync's .status is null when the child
+// could not be spawned (r.error set — e.g. a broken global `pmos` shim after a partial uninstall;
+// onPath only proves resolvability, not runnability) or died on a signal. The old `.status || 0`
+// coerced both to exit 0, so a launch that launched nothing reported success to whatever invoked it.
+function launchStatus(r) {
+  if (r.error) {
+    console.error(`create-pmos: ERROR — launch failed: ${r.error.message}`);
+    return 1;
+  }
+  return r.status === null ? 1 : r.status;
+}
+
 function launchWorkspace(targetDir, viaGlobal) {
   if (viaGlobal && onPath('pmos')) {
     // `pmos` owns the whole lifecycle: start the server detached, open the browser on this repo.
-    return spawnSync('pmos', [targetDir], { stdio: 'inherit', shell: process.platform === 'win32' }).status || 0;
+    return launchStatus(spawnSync('pmos', [targetDir], { stdio: 'inherit', shell: process.platform === 'win32' }));
   }
   // Fallback: run the bundled launcher directly. Its ensureInstalled() pulls runtime deps on first run.
   console.log('Launching from the bundled copy (the `pmos` command is not on PATH).');
-  return spawnSync(process.execPath, [path.join(WORKSPACE, 'bin', 'pmos.js'), targetDir],
-    { stdio: 'inherit', env: process.env }).status || 0;
+  return launchStatus(spawnSync(process.execPath, [path.join(WORKSPACE, 'bin', 'pmos.js'), targetDir],
+    { stdio: 'inherit', env: process.env }));
 }
 
 // Compare two dotted numeric versions. Returns -1 if a<b, 0 if equal, 1 if a>b. Missing/non-numeric

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "create-pmos",
-  "version": "0.34.0",
+  "version": "0.36.0",
   "description": "Scaffold the PMOS portable kit — run the PMOS product-management workflow (initiative → lane-fit spec → contract → build → three gates, with work-type lanes for features, research, maintenance, incidents, and content) entirely inside your own repo, with all state as git-versioned files. Includes `workflow-sweep`, which audits the state every board derives from and separates real defects from work you simply have not done. `create-pmos workspace` also installs + launches the local Terminal Workspace GUI. No database, no network, no telemetry.",
   "license": "MIT",
   "bin": {

package/payload/KIT-VERSION

@@ -1,3 +1,3 @@
-kernel-source: pmos@1313507
-assembled: 2026-08-07
+kernel-source: pmos@f04854e
+assembled: 2026-08-08
 mode: vendor

package/payload/okf/core/concepts/agent-harness.md

@@ -24,12 +24,12 @@ The harness includes:
 - **Knowledge** — the [OKF bundle](/okf/core/concepts/okf-governance.md): concepts, templates,
   playbooks.
 - **Skills** — procedural competence loaded on demand
-  ([SKILL-FORMAT](/skills/core/SKILL-FORMAT.md)).
+  ([SKILL-FORMAT](/skills/SKILL-FORMAT.md)).
 - **Tools / MCP** — the [ACI surface](/build-skills/MCP-SPEC.md) the agent acts through.
 - **Sandboxes** — where work runs safely.
 - **Orchestration** — sequencing, concurrency, routing.
 - **Guardrails** — the [three-tier eval gate](/okf/core/concepts/output-eval.md), kill conditions.
-- **Observability** — `agent_runs`, [discoveries.md](/discoveries.md), transcript review.
+- **Observability** — `agent_runs`, [discoveries.md](/state/discoveries.md), transcript review.
 
 **PMOS builds the harness, not the app.** Improving any harness element improves every future run —
 that is the leverage the [control plane](/okf/core/concepts/control-plane.md) exists to capture.

package/payload/okf/core/concepts/design-loop.md

@@ -29,12 +29,12 @@ An initiative with a user-facing surface runs four legs *inside* Build:
                                           build, web-ci)                library back)
 ```
 
-1. **Brief** — the [design-brief skill](/skills/core/design-brief.skill) writes the context handoff: what
+1. **Brief** — the [design-brief skill](/skills/design-brief.skill) writes the context handoff: what
    to design and the hard, presentation-only constraints, each surface tied to a real data source. It is to
-   design what the [PRD](/skills/core/prd.skill) is to build.
+   design what the [PRD](/skills/prd.skill) is to build.
 2. **Design** — the design is made in an external design surface (PMOS-self uses claude.ai/design;
    any capable design surface fits — external to this repo).
-3. **Reconcile + build** — the [design-reconcile skill](/skills/core/design-reconcile.skill) classifies
+3. **Reconcile + build** — the [design-reconcile skill](/skills/design-reconcile.skill) classifies
    every surface against **the product's capability ledger** and the PM signs off before any code; then the
    design is built into the product's app on its design system, verified by the product's CI. (For PMOS's own
    app the ledger is [web-capability-ledger.md](/planning/web-capability-ledger.md), the system is D38, and
@@ -98,8 +98,8 @@ time such that the board hides real "designed-but-not-built" state, the lighter
 
 The loop is a **method**; PMOS's own app is just its first instance. What is **shared** vs **per-product** (D50):
 
-- **Shared, authored once, read by every PM** — this concept, the [design-brief](/skills/core/design-brief.skill)
-  and [design-reconcile](/skills/core/design-reconcile.skill) skills, the
+- **Shared, authored once, read by every PM** — this concept, the [design-brief](/skills/design-brief.skill)
+  and [design-reconcile](/skills/design-reconcile.skill) skills, the
   [design-brief template](/okf/core/templates/design-brief-template.md), and the **generic dimensions** of the
   [design-fidelity rubric](/planning/evals/design-fidelity-eval-rubric.md). These live in PMOS-self's knowledge
   and are already reachable by tenants (repo-managed OKF rows are shared-read, D37; the `skills` table is

package/payload/okf/core/concepts/eval-driven-pm.md

@@ -52,7 +52,7 @@ this the right thing?* — and on whether the rubric itself was right.
 ## How it shapes the loop
 
 - **Spec stage** authors the rubric alongside the PRD — the
-  [eval-rubric skill](/skills/core/eval-rubric.skill) and [prd skill](/skills/core/prd.skill) are
+  [eval-rubric skill](/skills/eval-rubric.skill) and [prd skill](/skills/prd.skill) are
   paired for this reason.
 - **Contract stage** turns the rubric into agreed, testable done-criteria the generator and evaluator
   both sign off on before work starts.

package/payload/okf/core/concepts/okr-anti-optimism.md

@@ -36,7 +36,7 @@ report the more pessimistic reading and make the uncertainty explicit.
 4. **Name the gap to intent.** Where the work has bent away from the parent objective, say so
    explicitly, even if every individual run passed.
 5. **Surface friction.** Anything that fought the harness goes to
-   [discoveries.md](/discoveries.md) tagged `#friction`, not buried in a success summary.
+   [discoveries.md](/state/discoveries.md) tagged `#friction`, not buried in a success summary.
 
 ## Why this is structural, not cultural
 

package/payload/okf/core/concepts/okr-tree.md

@@ -29,8 +29,8 @@ Objective (qualitative, directional — the "what we want to be true")
 - **Key Result** — a quantitative, measurable outcome that proves the objective is being met.
   Each objective has 2–5 KRs. *"Human correction rate per agent run < 0.5 by end of Q3."*
 - **Initiative** — a concrete body of work expected to move one or more KRs. Each initiative owns a
-  PRD ([prd skill](/skills/core/prd.skill)) and an eval rubric
-  ([eval-rubric skill](/skills/core/eval-rubric.skill)).
+  PRD ([prd skill](/skills/prd.skill)) and an eval rubric
+  ([eval-rubric skill](/skills/eval-rubric.skill)).
 - **Agent run** — the leaf: one execution against an initiative's pre-run contract, logged to
   `agent_runs`.
 

package/payload/okf/core/concepts/output-eval.md

@@ -10,7 +10,7 @@ timestamp: 2026-06-28
 
 **Situating context:** Every turn of the [SDLC loop](/okf/core/concepts/sdlc-loop.md) ends by
 judging what an agent produced. This concept defines the three-tier gate that judgment runs through,
-so the [eval-rubric skill](/skills/core/eval-rubric.skill), AGENTS.md, and the PM all use the same
+so the [eval-rubric skill](/skills/eval-rubric.skill), AGENTS.md, and the PM all use the same
 vocabulary. It informs where automation stops and human judgment begins.
 
 ## The three tiers
@@ -72,7 +72,7 @@ green Review Gate auto-accept — removes the only structural defense against dr
   call sequences.
 - **Weighted, with partial credit:** the Review Gate scores weighted dimensions and totals against a
   threshold; it is not all-or-nothing. (See the
-  [eval-rubric skill](/skills/core/eval-rubric.skill) and
+  [eval-rubric skill](/skills/eval-rubric.skill) and
   [template](/okf/core/templates/eval-rubric-template.md).)
 
 ## Relationship to autonomy

package/payload/okf/core/concepts/sdlc-loop.md

@@ -26,7 +26,7 @@ PMOS drives a closed lifecycle. Each turn of the loop moves one initiative throu
                                                              Gate)
 ```
 
-1. **Spec** — the initiative gets a PRD via the [prd skill](/skills/core/prd.skill). The PRD says
+1. **Spec** — the initiative gets a PRD via the [prd skill](/skills/prd.skill). The PRD says
    what to build and why, anchored to a KR in the [OKR tree](/okf/core/concepts/okr-tree.md) —
    or, for maintenance work, a declared [health budget](/okf/core/concepts/health-anchors.md) (D59).
 2. **Contract** — generator and evaluator agree a **pre-run contract**: testable done-criteria,

package/payload/okf/core/concepts/self-refining-harness.md

@@ -14,7 +14,7 @@ transcripts, `eval-calibration.md`). What it lacked was a named pattern — and
 that evidence back into harness improvements. This concept names it. It was extracted from the
 2026-07-01 "Agent = Model + Harness" research (which found the field converging on *harness engineering*
 as its own discipline, and "agents improving their own tools" as its highest-leverage loop) and it
-informs the [harness-refine skill](/build-skills/harness-refine.skill), [D47](/okf/products/pmos/adr/d47-self-refining-harness.md),
+informs the [harness-refine skill](/skills/harness-refine.skill), [D47](/okf/products/pmos/adr/d47-self-refining-harness.md),
 and the [D23](/okf/products/pmos/adr/d23-quarterly-cadence.md) quarterly harness review.
 
 ## The pattern
@@ -36,7 +36,7 @@ one level up — applied to the harness itself:
 1. **Detect** — a deterministic signal shows a gap: a `v_friction_recurrence` theme with no resolving
    decision, a recurring `evals` defect class, a persistent `eval-calibration.md` disagreement, a
    context-budget signal.
-2. **Draft** — the [harness-refine skill](/build-skills/harness-refine.skill) reads that evidence and
+2. **Draft** — the [harness-refine skill](/skills/harness-refine.skill) reads that evidence and
    drafts concrete, **evidence-cited** edits, each mapped to one artifact: an MCP tool description
    ([ACI](/build-skills/MCP-SPEC.md), D19), a skill body, an OKF concept, or an `AGENTS.md` heuristic.
    It **prunes** as readily as it adds (D23 / [prune the scaffolding](/PRINCIPLES.md)).
@@ -73,5 +73,5 @@ sources.
 - [control-plane](/okf/core/concepts/control-plane.md) — why PMOS edits the harness, not the app.
 - [output-eval](/okf/core/concepts/output-eval.md) — the Acceptance Gate this loop submits to.
 - [watermelon-flag](/okf/core/concepts/watermelon-flag.md) — the L4 failure mode the L3 bounds prevent.
-- Skill: [harness-refine](/build-skills/harness-refine.skill). Decision:
+- Skill: [harness-refine](/skills/harness-refine.skill). Decision:
   [D47](/okf/products/pmos/adr/d47-self-refining-harness.md).

package/payload/okf/core/templates/content-brief-template.md

@@ -10,7 +10,7 @@ timestamp: 2026-07-18
 
 **Situating context:** Authored with [D58](/okf/products/pmos/adr/d58-builder-profiles.md) as the
 content lane's spec artifact — the contract copy work is drafted and judged against. The
-[copywriting skill](/skills/domains/copywriting.skill) instantiates it. A rubric is recommended,
+[copywriting skill](/skills/copywriting.skill) instantiates it. A rubric is recommended,
 not required (PM call at D58 ratification); the optional section below is the recommended shape.
 
 > Usage: replace every `<…>` placeholder. An empty audience or an unsourced claim list means the

package/payload/okf/core/templates/design-brief-template.md

@@ -9,7 +9,7 @@ timestamp: 2026-07-01
 # Design Brief Template
 
 **Situating context:** This template is the canonical scaffold the
-[design-brief skill](/skills/core/design-brief.skill) instantiates — leg 1 of the
+[design-brief skill](/skills/design-brief.skill) instantiates — leg 1 of the
 [design loop](/okf/core/concepts/design-loop.md). It is to design what the
 [PRD template](/okf/core/templates/prd-template.md) is to build: the shared definition before work starts.
 Copy the block into `/planning/design/<slug>-design-brief.md` and fill it; the copy is an operational

package/payload/okf/core/templates/eval-rubric-template.md

@@ -9,7 +9,7 @@ timestamp: 2026-06-28
 # Eval Rubric Template
 
 **Situating context:** This template was authored as the canonical scaffold the
-[eval-rubric skill](/skills/core/eval-rubric.skill) instantiates, so every rubric uses the same
+[eval-rubric skill](/skills/eval-rubric.skill) instantiates, so every rubric uses the same
 weighted-scoring math and carries its sprint contract. Copy the block into
 `/planning/evals/<slug>-eval-rubric.md` and fill it; the copy is operational and does not live in the
 OKF bundle.

package/payload/okf/core/templates/log.md

@@ -3,7 +3,7 @@
 OKF reserved file (no frontmatter). Change history, newest first, ISO 8601 dates.
 
 ## 2026-07-01
-- Added `design-brief-template` — the design-brief scaffold (leg 1 of the design loop, D48), instantiated by the [design-brief skill](/skills/core/design-brief.skill). Distilled from the first instance, `planning/design-brief.md`.
+- Added `design-brief-template` — the design-brief scaffold (leg 1 of the design loop, D48), instantiated by the [design-brief skill](/skills/design-brief.skill). Distilled from the first instance, `planning/design-brief.md`.
 
 ## 2026-06-29
 - Creation — per-folder index/log added (OKF recursive progressive disclosure; D29 reversal direction).

package/payload/okf/core/templates/prd-template.md

@@ -9,7 +9,7 @@ timestamp: 2026-06-28
 # PRD Template
 
 **Situating context:** This template was authored as the canonical scaffold the
-[prd skill](/skills/core/prd.skill) instantiates, so every PRD in PMOS has the same shape and the
+[prd skill](/skills/prd.skill) instantiates, so every PRD in PMOS has the same shape and the
 same anchoring discipline. Copy the block below into `/planning/prd/<slug>-prd.md` and fill it; the
 copy is an operational instance and does not live in the OKF bundle.
 

package/payload/okf/core/templates/spike-brief-template.md

@@ -11,7 +11,7 @@ timestamp: 2026-07-17
 **Situating context:** Authored with [D56](/okf/products/pmos/adr/d56-initiative-lanes.md) as the
 discovery lane's spec artifact, resolving the eval-first paradox the SDLC stress test named: a
 prototype exists to *discover* its own done-criteria, so the lane contracts on a **question and a
-timebox**, not on requirements. The [discovery skill](/skills/core/discovery.skill) instantiates
+timebox**, not on requirements. The [discovery skill](/skills/discovery.skill) instantiates
 this template. Copy the block below into `/planning/prd/<id>-spike.md` and fill; the copy is an
 operational instance, not OKF knowledge.
 

package/payload/scripts/frontier.sh

@@ -18,8 +18,12 @@ DIR="${1:-state/tickets}"
 # --- field readers (frontmatter only: stop at the closing --- fence) ----------------------------
 fm() { awk 'NR>1 && /^---[[:space:]]*$/{exit} {print}' "$1"; }   # lines between the opening and closing ---
 field() { printf '%s\n' "$1" | sed -n "s/^$2:[[:space:]]*//p" | head -1 | sed 's/^["'\'']//;s/["'\'']$//;s/[[:space:]]*$//'; }
-# blocked_by is an inline list "[\"002\", \"005\"]"; extract the bare ids (digits) in order.
-blockers() { printf '%s\n' "$1" | sed -n 's/^blocked_by:[[:space:]]*//p' | head -1 | grep -oE '[0-9]+' || true; }
+# blocked_by is an inline list "[\"002\", \"005\"]"; extract the bare ids in order. Ids are NOT
+# assumed numeric — idof below derives slug ids (an explicit `id:` field, or a digitless filename),
+# and a digits-only extraction silently DROPPED every such blocker, listing a blocked ticket as
+# workable (fail-open). Every non-empty token survives instead: a token that matches no done ticket
+# keeps the ticket off the frontier, so a malformed or unknown blocker fails CLOSED.
+blockers() { printf '%s\n' "$1" | sed -n 's/^blocked_by:[[:space:]]*//p' | head -1 | sed 's/[][]//g' | tr ',' '\n' | sed 's/["'\'']//g' | tr -d ' \t\r' | grep -v '^$' || true; }
 # derive an id: explicit `id:` field, else the leading NNN of the filename.
 idof() { local i; i="$(field "$1" id)"; [ -n "$i" ] && { printf '%s' "$i"; return; }; basename "$2" | grep -oE '^[0-9]+' || basename "$2" .md; }
 

package/payload/scripts/gate-initiative.sh

@@ -49,9 +49,30 @@ printf '%s' "$id" | grep -qE '^[a-z0-9][a-z0-9._-]*$' || fail "Initiative id '$i
 init_file="$KIT_ROOT/state/initiatives/$id.md"
 [ -f "$init_file" ] || fail "No initiative file at $init_file. Create it first — an initiative must exist before its PR. This is the loop rule the gate enforces."
 
-frontval() { # frontval <key> — first 'key: value' line in the file, value trimmed.
-  # `|| true`: an absent key must fall through to the explicit fail() below, not abort silently.
-  { grep -m1 -E "^$1:[[:space:]]*" "$init_file" || true; } | sed -E "s/^$1:[[:space:]]*//" | tr -d '\r' | sed -E 's/[[:space:]]+$//'
+frontval() { # frontval <key> — the key's value from the LEADING frontmatter block, trimmed.
+  # The LEADING block only, mirroring workflow-state.js / file-mode-parse.js / metrics.py: the gate
+  # used to grep the whole file, so a body line starting `parent_kr: ...` (a pasted example) could
+  # fake the declaration the gate exists to check. Same rules as the canonical parsers: an opening
+  # --- fence (before any key) is skipped and the next --- ends the block; blank lines are skipped;
+  # the first non-key line ends the block (fenceless records, e.g. the kit-ci smoke fixture, keep
+  # parsing from line 1); first value wins; ONE pair of surrounding double quotes is stripped. An
+  # absent key prints nothing and falls through to the explicit fail() below.
+  awk -v want="$1" '
+    /^---/ { if (fence || seen) exit; fence = 1; next }
+    /^[[:space:]]*$/ { next }
+    {
+      i = index($0, ":")
+      key = (i > 0) ? substr($0, 1, i - 1) : ""
+      if (key !~ /^[a-z_]+$/) exit
+      seen = 1
+      if (key != want) next
+      val = substr($0, i + 1)
+      gsub(/\r/, "", val)
+      gsub(/^[[:space:]]+|[[:space:]]+$/, "", val)
+      if (length(val) >= 2 && val ~ /^".*"$/) val = substr(val, 2, length(val) - 2)
+      print val
+      exit
+    }' "$init_file"
 }
 prd="$(frontval prd_path)"
 rubric="$(frontval rubric_path)"

package/payload/scripts/metrics.py

@@ -56,14 +56,30 @@ def leading_frontmatter(path):
     body line, the loop breaks, and every key below it is dropped. Agreement with the JS parser was
     never sufficient to catch this — both carried the same wrong regex, so the fixture compared two
     identical mistakes and passed. See kit/fixtures/state/initiatives/epsilon-empty-value.md.
+
+    The closing --- fence is TERMINAL, never skipped: stepping over it absorbed a key-shaped first
+    body line into the block (zeta-fence-fake.md pins this). Fenceless files parse from line 1 and
+    stop at the first non-key line. First value wins, and ONE pair of surrounding double quotes is
+    stripped, exactly as the JS parsers do (eta-quoted-value.md pins the quoted case).
     """
     fm = {}
+    fence_open = False
     with open(path, encoding="utf-8") as fh:
-        for line in fh:
-            m = re.match(r"^([a-z_]+):\s*(.*?)\s*$", line.rstrip("\n"))
+        for raw in fh:
+            line = raw.rstrip("\n")
+            if line.startswith("---"):
+                # Only the FIRST ---, before any key, opens the block; every later --- ends it.
+                if fence_open or fm:
+                    break
+                fence_open = True
+                continue
+            m = re.match(r"^([a-z_]+):\s*(.*?)\s*$", line)
             if m:
-                fm.setdefault(m.group(1), m.group(2))
-            elif line.strip() and not line.startswith("---"):
+                val = m.group(2)
+                if len(val) >= 2 and val.startswith('"') and val.endswith('"'):
+                    val = val[1:-1]
+                fm.setdefault(m.group(1), val)
+            elif line.strip():
                 break  # frontmatter is the LEADING block only
     return fm
 

package/payload/scripts/read-jsonl.js

@@ -0,0 +1,53 @@
+'use strict';
+/*
+ * scripts/read-jsonl.js — THE JSONL reader for the scripts/ checkers and the workflow engine.
+ *
+ * Six near-identical readers grew here, with three different torn-line contracts, and the repo's
+ * own comments record the cost twice: a silently-dropped torn line re-opened a verified entry
+ * (check-verify-queue.js) and made an accepted initiative read as never accepted (workflow-state.js).
+ * Each new checker re-decided the torn-line question from scratch and got it wrong at least twice.
+ * One reader, one contract — the strictest one found:
+ *
+ *   - a torn line is DROPPED, never guessed at — a truncated row parsed "as far as it goes" would
+ *     fabricate a record nobody wrote;
+ *   - and it is REPORTED, never swallowed: every defect comes back with its file, 1-based line,
+ *     parse reason and a bounded slice of the text, so the caller can say out loud what it could
+ *     not read. "Assuming someone else counts is how nobody counts."
+ *   - a MISSING (or unreadable) file is its own state, distinguishable from an empty one — the
+ *     cadence checker needs the difference and a gate must not confuse "no ledger" with "no rows".
+ *
+ * Returns { rows, defects, torn, missing }: rows are the parsed objects in file order; defects is
+ * [{ file, line, reason, text }]; torn === defects.length, kept as a count because most callers
+ * only need the number; missing is true when the file could not be read at all (rows/defects empty).
+ *
+ * DELIBERATELY FLAT in scripts/, not scripts/lib/: the kit and the workspace bundle both vendor the
+ * engine's require closure by BASENAME into a flat directory (kit/build-kit.sh, scripts/
+ * vendor-workspace-engine.js), so a `./lib/` require would resolve in the checkout and break in
+ * every vendored copy — the exact incident class the scripts/lib/sink.js re-export note records.
+ * workspace/lib/file-mode-parse.js does NOT delegate here for the mirrored reason: the workspace
+ * bundle ships alone, cannot require across the bundle boundary, and its silent-drop reader is the
+ * locked mirror of kit/scripts/metrics.py (kit/fixtures/test_parsers.py holds that agreement).
+ *
+ * Zero dependencies, stock Node.
+ */
+const fs = require('fs');
+
+function readJsonl(file) {
+  let text;
+  try { text = fs.readFileSync(file, 'utf8'); }
+  catch { return { rows: [], defects: [], torn: 0, missing: true }; }
+  const rows = [];
+  const defects = [];
+  const lines = text.split('\n');
+  for (let i = 0; i < lines.length; i++) {
+    const t = lines[i].trim();
+    if (!t) continue;
+    try { rows.push(JSON.parse(t)); }
+    catch (e) {
+      defects.push({ file, line: i + 1, reason: String((e && e.message) || 'not valid JSON').slice(0, 120), text: t.slice(0, 120) });
+    }
+  }
+  return { rows, defects, torn: defects.length, missing: false };
+}
+
+module.exports = { readJsonl };