@gobing-ai/spur 0.3.93 → 0.3.95

package/plugins/sp/scripts/validate-commands.ts

@@ -66,7 +66,7 @@ function parseCommand(filePath: string, name: string): ParsedCommand {
     // Extract frontmatter between --- delimiters
     const fmMatch = raw.match(/^---\n([\s\S]*?)\n---/);
     const body = fmMatch ? raw.slice(fmMatch[0].length).trim() : raw;
-    const fm = fmMatch ? fmMatch[1] : '';
+    const fm = fmMatch?.[1] ?? '';
 
     const description = extractYamlField(fm, 'description');
     const argumentHint = extractYamlField(fm, 'argument-hint');
@@ -124,9 +124,9 @@ function parseCommand(filePath: string, name: string): ParsedCommand {
 /** Extract a plain YAML string field (handles quoted and unquoted). */
 function extractYamlField(fm: string, key: string): string | undefined {
     const re = new RegExp(`^${key}:\\s*(.+)$`, 'm');
-    const m = fm.match(re);
-    if (!m) return undefined;
-    const value = m[1].trim();
+    const m = fm.match(re)?.[1];
+    if (m === undefined) return undefined;
+    const value = m.trim();
     // Unquote double-quoted YAML scalars
     if (value.startsWith('"') && value.endsWith('"')) {
         return value.slice(1, -1).replace(/\\"/g, '"');
@@ -137,9 +137,9 @@ function extractYamlField(fm: string, key: string): string | undefined {
 /** Extract a YAML flow list ["A", "B"]. */
 function extractYamlList(fm: string, key: string): string[] | undefined {
     const re = new RegExp(`^${key}:\\s*\\[(.*)\\]$`, 'm');
-    const m = fm.match(re);
-    if (!m) return undefined;
-    return [...m[1].matchAll(/"([^"]*)"/g)].map((q) => q[1]);
+    const m = fm.match(re)?.[1];
+    if (m === undefined) return undefined;
+    return [...m.matchAll(/"([^"]*)"/g)].map((q) => q[1] ?? '');
 }
 
 const ALLOWED_HEADINGS: Record<string, true> = { Usage: true, Implementation: true };
@@ -262,7 +262,7 @@ function checkTargetResolution(cmd: ParsedCommand, skillsDir: string, root: stri
 
     // Extract ## Implementation section content
     const implPart = cmd.body.split(/^## Implementation\n/m)[1] ?? '';
-    const implSection = implPart.split(/\n## (?!Usage)/m)[0].trim();
+    const implSection = (implPart.split(/\n## (?!Usage)/m)[0] ?? '').trim();
 
     // Skill() calls: Skill(skill="sp:NAME",
     // The name charset must stay at least as wide as any name a skill directory can
@@ -270,7 +270,7 @@ function checkTargetResolution(cmd: ParsedCommand, skillsDir: string, root: stri
     // would pass a genuinely unresolved reference rather than report it.
     const skillCalls = [...implSection.matchAll(/Skill\(skill="sp:([a-z][a-z0-9-]*)"/g)];
     for (const ref of skillCalls) {
-        const skillName = ref[1];
+        const skillName = ref[1] ?? '';
         const skillPath = join(skillsDir, skillName, 'SKILL.md');
         if (!existsSync(skillPath)) {
             violations.push({
@@ -304,7 +304,7 @@ function checkTargetResolution(cmd: ParsedCommand, skillsDir: string, root: stri
     // Procedure reference + anchor — markdown links with anchors in Implementation
     const anchorRefs = [...implSection.matchAll(/\[([^\]]*)\]\(([^)]*#([^)]*))\)/g)];
     for (const ref of anchorRefs) {
-        const filePart = ref[2].split('#')[0];
+        const filePart = (ref[2] ?? '').split('#')[0] ?? '';
         const anchor = ref[3];
         // Resolve relative paths from commands/ directory
         const resolved = resolve(join(root, 'plugins', 'sp', 'commands'), filePart);
@@ -414,7 +414,7 @@ function parseMarkdownTable(section: string): MarkdownTable | null {
             .replace(/\|$/, '')
             .split(/(?<!\\)\|/)
             .map((c) => c.trim().replace(/\\\|/g, '|'));
-    const header = split(tableLines[0]);
+    const header = split(tableLines[0] ?? '');
     // Skip the separator row (--- cells).
     const dataRows = tableLines
         .slice(1)
@@ -430,7 +430,7 @@ function parseMarkdownTable(section: string): MarkdownTable | null {
 function extractHintTokens(hint: string): { positionals: string[]; flags: string[] } {
     const positionals = [...hint.matchAll(/<[^>]+>/g)].map((m) => m[0]);
     // Flags: capture the `--flag` literal, ignoring the value placeholder that follows.
-    const flags = [...hint.matchAll(/(--[a-z][a-z0-9-]*)/g)].map((m) => m[1]);
+    const flags = [...hint.matchAll(/(--[a-z][a-z0-9-]*)/g)].map((m) => m[1] ?? '');
     // Deduplicate while preserving order.
     return {
         positionals: [...new Set(positionals)],

package/plugins/sp/scripts/validate-flag-contracts.ts

@@ -71,7 +71,7 @@ function sectionBetween(raw: string, startRe: RegExp, nextHeading: '## ' | '###
 function extractFlags(text: string): Set<string> {
     const flags = new Set<string>();
     for (const m of text.matchAll(/(--[a-z][a-z-]*)/g)) {
-        flags.add(m[1]);
+        flags.add(m[1] ?? '');
     }
     return flags;
 }
@@ -80,7 +80,7 @@ function extractFlags(text: string): Set<string> {
 function deprecatedFlagSet(hint: string): Set<string> {
     const out = new Set<string>();
     for (const m of hint.matchAll(/(--[a-z][a-z-]*)\s*\(deprecated\)/g)) {
-        out.add(m[1]);
+        out.add(m[1] ?? '');
     }
     return out;
 }
@@ -104,7 +104,7 @@ function glossaryEntries(glossaryRaw: string): GlossaryEntry[] {
     const entryRe = /^### `(--[a-z][a-z-]*)[^\n]*$/gm;
     for (const m of glossaryRaw.matchAll(entryRe)) {
         const flag = m[1];
-        if (m.index === undefined) continue;
+        if (flag === undefined || m.index === undefined) continue;
         const bodyStart = m.index + m[0].length;
         const next = glossaryRaw.indexOf('\n### ', bodyStart);
         const body = next === -1 ? glossaryRaw.slice(bodyStart) : glossaryRaw.slice(bodyStart, next);
@@ -132,8 +132,8 @@ function glossaryDeclaringClaims(glossaryRaw: string): Map<string, Set<string>>
     for (const { flag, firstParagraph } of glossaryEntries(glossaryRaw)) {
         const names = new Set<string>();
         for (const paren of firstParagraph.matchAll(/\(([^)]*)\)/g)) {
-            for (const n of paren[1].matchAll(/`(dev-[a-z-]+)`/g)) {
-                names.add(n[1]);
+            for (const n of (paren[1] ?? '').matchAll(/`(dev-[a-z-]+)`/g)) {
+                names.add(n[1] ?? '');
             }
         }
         if (names.size >= 2) claims.set(flag, names);
@@ -210,7 +210,7 @@ function commandTableDefaults(commandRaw: string): Map<string, string> {
     if (!start || start.index === undefined) return out;
     const from = start.index + start[0].length;
     const next = commandRaw.indexOf('\n## ', from);
-    const table = (next === -1 ? commandRaw.slice(from) : commandRaw.slice(from, next)).split(/\n\s*\n/)[0];
+    const table = (next === -1 ? commandRaw.slice(from) : commandRaw.slice(from, next)).split(/\n\s*\n/)[0] ?? '';
     // Split on unescaped pipes only (same as validate-commands.ts parseMarkdownTable),
     // then unescape `\|` → `|` per cell.
     const split = (row: string): string[] =>
@@ -223,10 +223,11 @@ function commandTableDefaults(commandRaw: string): Map<string, string> {
         const trimmed = line.trim();
         if (!trimmed.startsWith('|')) continue;
         const cells = split(trimmed);
-        if (cells.length < 3 || cells[0].startsWith('---') || cells[0] === 'Flag') continue;
-        const flag = cells[0].replace(/`/g, '').split(/\s/)[0];
+        const first = cells[0] ?? '';
+        if (cells.length < 3 || first.startsWith('---') || first === 'Flag') continue;
+        const flag = first.replace(/`/g, '').split(/\s/)[0] ?? '';
         if (!flag.startsWith('--')) continue;
-        out.set(flag, cells[2].replace(/[`*]/g, '').trim());
+        out.set(flag, (cells[2] ?? '').replace(/[`*]/g, '').trim());
     }
     return out;
 }
@@ -237,7 +238,7 @@ function commandToOpHeader(opsRaw: string): Map<string, string> {
     const rowRe = /^\|\s*(\d+[a-z]?)\s*\|\s*([a-z-]+)\s*\|\s*`(dev-[a-z-]+)`/;
     for (const line of opsRaw.split('\n')) {
         const m = line.match(rowRe);
-        if (m) map.set(m[3], `${m[1]}. ${m[2]}`);
+        if (m) map.set(m[3] ?? '', `${m[1] ?? ''}. ${m[2] ?? ''}`);
     }
     return map;
 }
@@ -249,17 +250,18 @@ function opsSectionDefaults(sectionBody: string): Map<string, { value: string; c
         // Pair each default marker with the flag mentioned immediately before it on the line.
         const flags: Array<{ name: string; index: number }> = [];
         for (const m of line.matchAll(/`(--[a-z][a-z-]*)(?: <[^>]+>)?`/g)) {
-            flags.push({ name: m[1], index: m.index });
+            flags.push({ name: m[1] ?? '', index: m.index });
         }
         for (const m of line.matchAll(/\(default:?\s*\*?\*?`?([^`);]+?)\s*`?[;)]/g)) {
-            const value = m[1].replace(/[`*]/g, '').trim();
+            const value = (m[1] ?? '').replace(/[`*]/g, '').trim();
             const preceding = flags.filter((f) => f.index < m.index).pop();
             if (!preceding) continue;
             out.set(preceding.name, { value, claim: line.trim() });
         }
         // `X (default)` trailing form (e.g. "**`full`** (default):" in Modes prose).
         for (const m of line.matchAll(/`([^`]+)`\s*\(default\)/g)) {
-            out.set(m[1], { value: m[1], claim: line.trim() });
+            const v = m[1] ?? '';
+            out.set(v, { value: v, claim: line.trim() });
         }
     }
     return out;
@@ -369,11 +371,12 @@ export function extractValueBehaviorTable(
             .replace(/^\||\|$/g, '')
             .split('|')
             .map((c) => c.trim());
-        if (cells.length < 3 || cells[0].startsWith('---') || cells[0] === 'Value') continue;
-        const valueCell = cells[0].replace(/`/g, '').trim();
-        const value = valueCell.split(/\s/)[0];
+        const first = cells[0] ?? '';
+        if (cells.length < 3 || first.startsWith('---') || first === 'Value') continue;
+        const valueCell = first.replace(/`/g, '').trim();
+        const value = valueCell.split(/\s/)[0] ?? '';
         if (value !== 'inline' && value !== 'auto' && value !== '<name>') continue;
-        const behavior = normalizeSurfaceCell(cells[2]);
+        const behavior = normalizeSurfaceCell(cells[2] ?? '');
         out.set(value, { ...behavior, defaultWhenOmitted: /default when omitted/i.test(valueCell) });
     }
     return out.size === 3 ? out : null;
@@ -395,9 +398,10 @@ export function extractTriggerTable(crossCuttingRaw: string): string[] | null {
             .replace(/^\||\|$/g, '')
             .split('|')
             .map((c) => c.trim());
-        if (cells.length < 3 || cells[0].startsWith('---')) continue;
-        if (cells[0] === 'Trigger') continue;
-        rows.push(cells[0].replace(/\*\*/g, '').replace(/\.\s*$/, ''));
+        const first = cells[0] ?? '';
+        if (cells.length < 3 || first.startsWith('---')) continue;
+        if (first === 'Trigger') continue;
+        rows.push(first.replace(/\*\*/g, '').replace(/\.\s*$/, ''));
     }
     return rows.length >= 4 ? rows : null;
 }
@@ -414,8 +418,12 @@ function adrAgentClaims(adrRaw: string): Map<string, SurfaceBehavior> | null {
             conditional: false,
             defaultWhenOmitted: false,
         });
-        out.set('auto', { surfaces: new Set(['subprocess']), conditional: false });
-        out.set('<name>', { surfaces: new Set(['inline', 'subprocess']), conditional: true });
+        out.set('auto', { surfaces: new Set(['subprocess']), conditional: false, defaultWhenOmitted: false });
+        out.set('<name>', {
+            surfaces: new Set(['inline', 'subprocess']),
+            conditional: true,
+            defaultWhenOmitted: false,
+        });
         return out;
     }
     // Slice from the `## ADR-041` heading to the next `## ADR-` heading or end of input
@@ -429,7 +437,11 @@ function adrAgentClaims(adrRaw: string): Map<string, SurfaceBehavior> | null {
     const out = new Map<string, SurfaceBehavior>();
     // Rule sentence: "if the named executor is … inline; otherwise … subprocess" → <name> conditional.
     if (/the work happens inline;\s*otherwise it dispatches a subprocess/.test(section)) {
-        out.set('<name>', { surfaces: new Set(['inline', 'subprocess']), conditional: true });
+        out.set('<name>', {
+            surfaces: new Set(['inline', 'subprocess']),
+            conditional: true,
+            defaultWhenOmitted: false,
+        });
     }
     // Collapse mapping: `--subprocess` → `--agent <target>`. The target value's behavior is
     // claimed by the mapping itself — parse the target generically so a drifted mapping
@@ -438,12 +450,12 @@ function adrAgentClaims(adrRaw: string): Map<string, SurfaceBehavior> | null {
     // `--subprocess` is matched explicitly (`` `--subprocess` `` in the source).
     const subprocessTarget = section.match(/--subprocess`\s*→\s*`--agent ([a-z]+)/)?.[1];
     if (subprocessTarget === 'auto') {
-        out.set('auto', { surfaces: new Set(['subprocess']), conditional: false });
+        out.set('auto', { surfaces: new Set(['subprocess']), conditional: false, defaultWhenOmitted: false });
     } else if (subprocessTarget === 'inline') {
-        out.set('auto', { surfaces: new Set(['inline']), conditional: false });
+        out.set('auto', { surfaces: new Set(['inline']), conditional: false, defaultWhenOmitted: false });
     }
     if (/--inline`\s*→\s*`--agent inline/.test(section)) {
-        out.set('inline', { surfaces: new Set(['inline']), conditional: false });
+        out.set('inline', { surfaces: new Set(['inline']), conditional: false, defaultWhenOmitted: false });
     }
     return out.size === 0 ? null : out;
 }
@@ -637,8 +649,8 @@ export function checkAgentSsotIntegrity(files: ReadonlyMap<string, string>): Fla
         for (const block of tableBlocks) {
             const valuesFound = new Set<string>();
             for (const row of block) {
-                const firstCell = row.replace(/^\|/, '').split('|')[0].replace(/`/g, '').trim();
-                const token = firstCell.split(/\s/)[0].toLowerCase();
+                const firstCell = (row.replace(/^\|/, '').split('|')[0] ?? '').replace(/`/g, '').trim();
+                const token = (firstCell.split(/\s/)[0] ?? '').toLowerCase();
                 if (token === 'inline' || token === 'auto' || token === '<name>') {
                     valuesFound.add(token);
                 }
@@ -684,7 +696,7 @@ export function headingAnchors(raw: string): Set<string> {
     const anchors = new Set<string>();
     for (const line of raw.split('\n')) {
         const m = line.match(/^#{1,6}\s+(.*)$/);
-        if (m) anchors.add(headingSlug(m[1]));
+        if (m) anchors.add(headingSlug(m[1] ?? ''));
     }
     return anchors;
 }
@@ -704,7 +716,7 @@ export function checkSsotAnchorsResolve(crossCuttingRaw: string, files: Readonly
     for (const [filename, raw] of files) {
         const seen = new Set<string>();
         for (const m of raw.matchAll(/cross-cutting\.md#([\w-]+)/g)) {
-            const anchor = m[1];
+            const anchor = m[1] ?? '';
             if (anchors.has(anchor) || seen.has(anchor)) continue;
             seen.add(anchor);
             violations.push({

package/plugins/sp/scripts/verify-answer-lint.ts

@@ -42,7 +42,8 @@ function usage(): never {
 }
 
 function defaultSpurBin(): string {
-    if (getEnvVar('SPUR_BIN')) return getEnvVar('SPUR_BIN');
+    const fromEnv = getEnvVar('SPUR_BIN');
+    if (fromEnv) return fromEnv;
     const local = fileURLToPath(new URL('../../../apps/cli/src/index.ts', import.meta.url));
     if (existsSync(local)) return `bun ${local}`;
     return 'spur';
@@ -55,6 +56,7 @@ function parseArgs(argv: string[]): { wbs: string; answer: string; spurBin: stri
     let i = 0;
     while (i < argv.length) {
         const arg = argv[i];
+        if (arg === undefined) break;
         if (arg === '--spur-bin') {
             spurBin = argv[i + 1] ?? defaultSpurBin();
             i += 2;

package/plugins/sp/scripts/workflow-step-profile.mjs

@@ -176,8 +176,9 @@ function formatStepProfileHuman(profile) {
 `;
 }
 function defaultSpurBin() {
-  if (getEnvVar("SPUR_BIN"))
-    return getEnvVar("SPUR_BIN");
+  const fromEnv = getEnvVar("SPUR_BIN");
+  if (fromEnv)
+    return fromEnv;
   const local = fileURLToPath(new URL("../../../apps/cli/src/index.ts", import.meta.url));
   if (existsSync(local))
     return `bun ${local}`;
@@ -196,6 +197,8 @@ function parseStepProfileCliArgs(argv) {
   let help = false;
   for (let i = 0;i < argv.length; i++) {
     const a = argv[i];
+    if (a === undefined)
+      continue;
     if (a === "--help" || a === "-h")
       help = true;
     else if (a === "--json")

package/plugins/sp/scripts/workflow-step-profile.ts

@@ -322,7 +322,8 @@ export interface StepProfileCliArgs {
  * --spur-bin > SPUR_BIN > monorepo-local CLI entry > PATH `spur`.
  */
 export function defaultSpurBin(): string {
-    if (getEnvVar('SPUR_BIN')) return getEnvVar('SPUR_BIN');
+    const fromEnv = getEnvVar('SPUR_BIN');
+    if (fromEnv) return fromEnv;
     const local = fileURLToPath(new URL('../../../apps/cli/src/index.ts', import.meta.url));
     if (existsSync(local)) return `bun ${local}`;
     return 'spur';
@@ -343,6 +344,7 @@ export function parseStepProfileCliArgs(argv: string[]): StepProfileCliArgs {
 
     for (let i = 0; i < argv.length; i++) {
         const a = argv[i];
+        if (a === undefined) continue;
         if (a === '--help' || a === '-h') help = true;
         else if (a === '--json') json = true;
         else if (a === '--last') last = positiveInt(argv[++i], last);

package/plugins/sp/scripts/wrapup-drift-probe.mjs

@@ -57,6 +57,8 @@ function changedPathsOf(section) {
   const paths = [];
   for (const match of section.matchAll(CHANGE_ENTRY)) {
     const path = match[1];
+    if (path === undefined)
+      continue;
     if (/\s/.test(path))
       continue;
     if (!path.includes("/") && !path.includes("."))
@@ -73,7 +75,8 @@ var SURFACE_MATCHERS = DOC_OWNED_SURFACES.map((glob) => ({ glob, regex: surfaceR
 function driftReasonForPath(path) {
   if (CORPUS_PREFIXES.some((prefix) => path.startsWith(prefix)))
     return null;
-  if (path.includes("/") && !KNOWN_TOP_LEVEL.has(path.split("/")[0])) {
+  const top = path.split("/")[0];
+  if (path.includes("/") && top !== undefined && !KNOWN_TOP_LEVEL.has(top)) {
     return "new top-level workspace directory";
   }
   const match = SURFACE_MATCHERS.find((entry) => entry.regex.test(path));
@@ -82,7 +85,7 @@ function driftReasonForPath(path) {
 function defaultSpurRunner(env, cwd) {
   const parts = (env.spurBin ?? "spur").split(/\s+/).filter((part) => part.length > 0);
   return (args) => {
-    const run = spawnSync(parts[0], [...parts.slice(1), ...args], {
+    const run = spawnSync(parts[0] ?? "spur", [...parts.slice(1), ...args], {
       cwd,
       encoding: "utf8",
       env: getEnvVars()

package/plugins/sp/scripts/wrapup-drift-probe.ts

@@ -125,6 +125,7 @@ export function changedPathsOf(section: string | null): string[] {
     const paths: string[] = [];
     for (const match of section.matchAll(CHANGE_ENTRY)) {
         const path = match[1];
+        if (path === undefined) continue;
         // Path-like: no whitespace and at least one separator or dotted suffix, so prose
         // backticks never masquerade as change-map entries.
         if (/\s/.test(path)) continue;
@@ -148,7 +149,8 @@ const SURFACE_MATCHERS = DOC_OWNED_SURFACES.map((glob) => ({ glob, regex: surfac
 /** 0944 R2 surface match: doc-owned glob, corpus exclusion, new top-level directory. */
 export function driftReasonForPath(path: string): string | null {
     if (CORPUS_PREFIXES.some((prefix) => path.startsWith(prefix))) return null;
-    if (path.includes('/') && !KNOWN_TOP_LEVEL.has(path.split('/')[0])) {
+    const top = path.split('/')[0];
+    if (path.includes('/') && top !== undefined && !KNOWN_TOP_LEVEL.has(top)) {
         return 'new top-level workspace directory';
     }
     const match = SURFACE_MATCHERS.find((entry) => entry.regex.test(path));
@@ -158,7 +160,7 @@ export function driftReasonForPath(path: string): string | null {
 function defaultSpurRunner(env: DriftProbeEnv, cwd?: string): SpurRunner {
     const parts = (env.spurBin ?? 'spur').split(/\s+/).filter((part) => part.length > 0);
     return (args: string[]): SpurShowResult => {
-        const run = spawnSync(parts[0], [...parts.slice(1), ...args], {
+        const run = spawnSync(parts[0] ?? 'spur', [...parts.slice(1), ...args], {
             cwd,
             encoding: 'utf8',
             env: getEnvVars(),

package/plugins/sp/scripts/wrapup-steps.mjs

@@ -172,6 +172,31 @@ function readFileSyncSafe(path) {
     return null;
   }
 }
+function verdictFromTestingSection(content) {
+  const heading = /^#{1,6}\s+Testing\s*$/m.exec(content);
+  if (!heading)
+    return null;
+  const level = heading[0].match(/^#+/)?.[0]?.length ?? 2;
+  const rest = content.slice(heading.index + heading[0].length);
+  const next = new RegExp(`^#{1,${level}}\\s+\\S`, "m").exec(rest);
+  const section = next ? rest.slice(0, next.index) : rest;
+  for (const line of section.split(`
+`)) {
+    const m = /^(?:-\s*|\*\*)?Verdict:\s*(PASS|PARTIAL|FAIL|UNKNOWN)\b/i.exec(line.trim());
+    if (m?.[1] !== undefined)
+      return m[1].toUpperCase();
+  }
+  return null;
+}
+function verdictOfArtifact(verdictPath) {
+  try {
+    const raw = jqPick(JSON.parse(readFileSync(verdictPath, "utf8")).verdict, "");
+    const text = jqText(raw);
+    return text.length > 0 ? text : null;
+  } catch {
+    return null;
+  }
+}
 function runMetrics(env, options = {}) {
   const cwd = options.cwd;
   const runId = env.__runId ?? "";
@@ -212,15 +237,13 @@ function runMetrics(env, options = {}) {
     const frontmatter = parsed.frontmatter !== null && typeof parsed.frontmatter === "object" ? parsed.frontmatter : {};
     const featureId = String(jqPick(frontmatter.feature_id, parsed.feature_id, ""));
     const status2 = String(jqPick(frontmatter.status, parsed.status, "unknown"));
-    let verdict = "UNKNOWN";
     const verdictPath = join(".spur", "run", `${wbs}-verdict.json`);
-    if (existsSync(abs(verdictPath))) {
-      try {
-        const raw = jqPick(JSON.parse(readFileSync(abs(verdictPath), "utf8")).verdict, "UNKNOWN");
-        const text = raw === "UNKNOWN" ? "UNKNOWN" : jqText(raw);
-        if (text.length > 0)
-          verdict = text;
-      } catch {}
+    const artifactVerdict = verdictOfArtifact(abs(verdictPath));
+    const trackedVerdict = verdictFromTestingSection(typeof parsed.content === "string" ? parsed.content : "");
+    const verdict = artifactVerdict ?? trackedVerdict ?? "UNKNOWN";
+    if (verdict === "UNKNOWN") {
+      process.stderr.write(`metrics-record: task ${wbs} has no certifying verdict \u2014 ${verdictPath}: ${artifactVerdict ?? "missing or carries none"}, tracked Testing: ${trackedVerdict ?? "no Verdict: line"} \u2014 recording UNKNOWN telemetry
+`);
     }
     const timestamp = new Date().toISOString().replace(/\.\d{3}Z$/, "Z");
     const row = { wbs, feature_id: featureId, status: status2, verdict, timestamp };
@@ -403,6 +426,7 @@ function main(argv, env = getEnvVars(), options = {}) {
 }
 export {
   writeRouteReason,
+  verdictFromTestingSection,
   taskStatusOf,
   spurCommand,
   runMetrics,

package/plugins/sp/scripts/wrapup-steps.ts

@@ -269,6 +269,49 @@ function readFileSyncSafe(path: string): string | null {
     }
 }
 
+/**
+ * Verdict from a task record's tracked `Testing` section — the durable copy that outlives a
+ * worktree teardown or clone (F93). Local copy of `parseVerdictLine` and of the `Testing` slice
+ * in `extractTestingSection` (`packages/app/src/services/task-record.ts`) because ADR-065 keeps
+ * plugin scripts on builtin and relative imports only; keep the three literals in step by hand
+ * (the parity guard in `plugins/sp/tests/wrapup-steps.test.ts` reads the canonical literals).
+ *
+ * The slice rule is the canonical one verbatim — `#{1,6}` heading, end at the next
+ * same-or-higher heading — so an `# Testing` or `#### Testing` heading, and an h4 subheading
+ * inside a `## Testing` section, read the same here and in the app. One deliberate difference:
+ * with no `Testing` heading at all this returns null instead of falling back to the whole
+ * document, so a `Verdict:` token in another section cannot be misread as this task's verdict.
+ */
+export function verdictFromTestingSection(content: string): string | null {
+    const heading = /^#{1,6}\s+Testing\s*$/m.exec(content);
+    if (!heading) return null;
+    const level = heading[0].match(/^#+/)?.[0]?.length ?? 2;
+    const rest = content.slice(heading.index + heading[0].length);
+    const next = new RegExp(`^#{1,${level}}\\s+\\S`, 'm').exec(rest);
+    const section = next ? rest.slice(0, next.index) : rest;
+    for (const line of section.split('\n')) {
+        // Line-anchored (optionally after `- ` bullet or `**` bold) so evidence text
+        // containing a mid-line "Verdict:" token cannot be misread as the section verdict.
+        const m = /^(?:-\s*|\*\*)?Verdict:\s*(PASS|PARTIAL|FAIL|UNKNOWN)\b/i.exec(line.trim());
+        if (m?.[1] !== undefined) return m[1].toUpperCase();
+    }
+    return null;
+}
+
+/**
+ * `verdict` field of a verdict artifact; null when the file is absent, unreadable, malformed or
+ * carries no usable verdict (jq `//` semantics: null, undefined and false all count as missing).
+ */
+function verdictOfArtifact(verdictPath: string): string | null {
+    try {
+        const raw = jqPick((JSON.parse(readFileSync(verdictPath, 'utf8')) as { verdict?: unknown }).verdict, '');
+        const text = jqText(raw);
+        return text.length > 0 ? text : null;
+    } catch {
+        return null;
+    }
+}
+
 export interface MetricsResult {
     status: 'PASS' | 'FAIL';
     statusFile: string;
@@ -323,17 +366,20 @@ export function runMetrics(env: WrapupStepsEnv, options: WrapupStepsOptions = {}
         const featureId = String(jqPick(frontmatter.feature_id, parsed.feature_id, ''));
         const status = String(jqPick(frontmatter.status, parsed.status, 'unknown'));
 
-        // jq `//` semantics: null and false count as missing; an empty string result stays UNKNOWN.
-        let verdict = 'UNKNOWN';
+        // The verdict artifact stays the first source (F93 R2); when it is gone — a worktree run
+        // fast-forwards and removes the tree, taking the gitignored artifact with it — the tracked
+        // `Testing` line `task record` already wrote into the task file is the durable copy. Honest
+        // UNKNOWN is the last resort, and it is telemetry, never proof of completion.
         const verdictPath = join('.spur', 'run', `${wbs}-verdict.json`);
-        if (existsSync(abs(verdictPath))) {
-            try {
-                const raw = jqPick(JSON.parse(readFileSync(abs(verdictPath), 'utf8')).verdict, 'UNKNOWN');
-                const text = raw === 'UNKNOWN' ? 'UNKNOWN' : jqText(raw);
-                if (text.length > 0) verdict = text;
-            } catch {
-                // unreadable verdict file keeps UNKNOWN telemetry
-            }
+        const artifactVerdict = verdictOfArtifact(abs(verdictPath));
+        const trackedVerdict = verdictFromTestingSection(typeof parsed.content === 'string' ? parsed.content : '');
+        const verdict = artifactVerdict ?? trackedVerdict ?? 'UNKNOWN';
+        if (verdict === 'UNKNOWN') {
+            // R3: name the task, the artifact path and the tracked state. An explicit UNKNOWN on
+            // either source is still an uncertified row, so it reports the same way as a miss.
+            process.stderr.write(
+                `metrics-record: task ${wbs} has no certifying verdict — ${verdictPath}: ${artifactVerdict ?? 'missing or carries none'}, tracked Testing: ${trackedVerdict ?? 'no Verdict: line'} — recording UNKNOWN telemetry\n`,
+            );
         }
 
         const timestamp = new Date().toISOString().replace(/\.\d{3}Z$/, 'Z');

package/plugins/sp/skills/code-verification/SKILL.md

@@ -259,17 +259,22 @@ the deterministic Testing writer `spur task record` (section authorship never ha
 
 ```bash
 # write .spur/run/<wbs>-verdict.json (shape in references/verdict-schema.md), then:
-# F96 residual sweep (observe-only): scan + fold BEFORE record, under every --fix mode.
+spur task record <wbs> --verdict-file .spur/run/<wbs>-verdict.json  # renders ## Testing, flips proven boxes
+# F96 residual sweep (observe-only): scan + fold AFTER record, under every --fix mode (0987).
 RESIDUAL=$(superskill script path sp residual-scan.mjs)
 node "$RESIDUAL" scan <wbs>
 node "$RESIDUAL" fold <wbs>
-spur task record <wbs> --verdict-file .spur/run/<wbs>-verdict.json  # renders ## Testing
+# Downgrade only: re-record so Testing carries the final (folded) verdict.
+[ "$(jq -r .verdict .spur/run/<wbs>-verdict.json)" = PASS ] \
+  || spur task record <wbs> --verdict-file .spur/run/<wbs>-verdict.json
 ```
 
-> **Residual fold (F96).** `scan` reads `.spur/run/` artifacts and writes `residuals.json` +
-> `residual-report.md`; `fold` rewrites the **just-written** verdict artifact's `residual-sweep`
-> check in place (PASS → PARTIAL when a blocking residual exists), so `record` transcribes the
-> downgraded verdict — a manual verify cannot certify what the pipeline would reject. Resolve the
+> **Residual fold (F96).** `scan` reads `.spur/run/` artifacts and the post-record task file and
+> writes `residuals.json` + `residual-report.md`; `fold` rewrites the verdict artifact's
+> `residual-sweep` check in place (PASS → PARTIAL when a blocking residual exists), and the
+> downgrade-only re-record transcribes it — a manual verify cannot certify what the pipeline would
+> reject. The sweep runs after `record` (same order as the pipeline, 0983/0987): before it, a
+> verdict-proven box is still unticked and a clean PASS would fold to PARTIAL. Resolve the
 > script via `superskill script path sp residual-scan.mjs`; shipped surfaces never reference
 > `plugins/sp/scripts/` directly (script-contract-check rule 4). When the task reaches `done`
 > through `--next`, run `residual-scan settle` (links follow-up tasks; best-effort).
@@ -322,7 +327,7 @@ Verdict: PASS
 ### SECUA Review
 | Priority | Dimension | Location | Finding |
 | --- | --- | --- | --- |
-| P4 | — | — | No P1–P3 findings; verify verdict PASS |
+| P4 | — | — | No findings (verify verdict PASS) |
 ```
 
 The per-requirement traceability table MUST use `| Req | Status | Evidence |` (exactly this header, no `R#`/`R`/`Requirement` variant — `Status` in column 2 — and no extra columns between Req and Status). The Acceptance Criteria table MUST use `| AC | Status | Evidence Type | Evidence |`.

package/plugins/sp/skills/code-verification/references/verdict-schema.md

@@ -118,6 +118,17 @@ fail — invalid status (`MET | PARTIAL | UNMET` for requirements;
 manual-review | llm-judge | n/a`, or a `+` compound), and empty evidence. Interrupted runs keep the
 rows that pass the lint and complete only the missing IDs on retry.
 
+**Feature-credited AC ids are narrower than the lint's accepted set (task 0966 finding).** A row keyed
+by a bare checklist ordinal (`AC1`) passes `verify-answer-lint` — it resolves to the task's AC
+checklist label — but `feature-check.rowMatchesScenario` credits a linked feature scenario only by the
+scenario's normalized title or its `AC-N` ordinal alias. Such a row therefore carries a PASS verdict
+while crediting no scenario, and the feature's `verifying → done` gate denies with
+`L4.scenario-unverified` / `L4.verdict-rows-match-no-scenario`. When the task has a linked feature,
+key each AC row `AC-N` (1-based scenario ordinal, hyphenated) or by the exact scenario title.
+A scenario-keyed row (`R3 — <title>`) ticks the task AC box whose line aliases that scenario
+(`- [ ] AC1 — R3 — <title>`) during `spur task record` (0996); a scenario key no AC line aliases
+ticks nothing, so alias every graduated scenario on its AC line.
+
 **Concrete anchors only (task 0804 R9).** An evidence anchor must be a concrete existing `file:line`
 (or `file:start-end`) path. A glob or directory summary (`src/services/*.ts`, `the retry
 classifiers in task-pipeline.yaml`) is not an anchor: expand it into the specific cited files/ranges

package/plugins/sp/skills/functional-review/SKILL.md

@@ -253,7 +253,7 @@ shape, e.g. `| Req | Status | Evidence |` alone, is structurally rejected and de
 ```markdown
 | Priority | Dimension | Location | Finding |
 | --- | --- | --- | --- |
-| P4 | — | — | No P1–P3 findings; functional verdict PASS |
+| P4 | — | — | No findings (functional verdict PASS) |
 
 | Req | Status | Evidence |
 | --- | --- | --- |

package/plugins/sp/skills/next-router/references/routing-table.md

@@ -113,7 +113,7 @@ token, deterministic).
 | C3 | A5/A6 when Testing empty/N/A **and** verify would fail for missing tests — only if prior implement claims code exists | Coverage/test signal: `bun test` fail attributed to task paths OR explicit "insufficient tests" in prior verify verdict artifact `.spur/run/<wbs>-verdict.json` | test fail / coverage gap | `/sp:dev-unit <wbs> --auto` | continue |
 | C4 | A3/A5/A6 when operator or task tags mention rules, OR `spur rule run` last report dirty in `.spur/` if present | `spur rule run` (default project preset) non-zero with findings | rule findings | **HITL STOP** — print rule summary; suggest `/sp:rule-scan` or `rule-add`/`rule-refine` (do not auto-author rules) | continue |
 | C5 | A6 only | Existing `.spur/run/<wbs>-verdict.json` with FAIL and findings pointing at coverage | verdict artifact | `/sp:dev-unit <wbs>` then re-verify on next invocation (`--once` friendly) | `/sp:dev-verify …` |
-| C6 | A4/A5 | `.spur/run/<wbs>-verdict.json` has a failing (PARTIAL/FAIL) `residual-sweep` check — the bounded remediation loop already ran and failed (F96) | folded verdict artifact | **HITL STOP** — print `.spur/run/<wbs>-residual-report.md` and the recovery command `/sp:dev-run <wbs>`; never auto-dispatch a fix (repeating the loop unattended burns quota without new information) | continue |
+| C6 | A4/A5 | `.spur/run/<wbs>-verdict.json` has a failing (PARTIAL/FAIL) `residual-sweep` check — residual blockers survived the pipeline sweep (F96) | folded verdict artifact | **HITL STOP** — print `.spur/run/<wbs>-residual-report.md` and the recovery command `/sp:dev-run <wbs>`; never auto-dispatch a fix (repeating the loop unattended burns quota without new information) | continue |
 
 **C-row precedence note (F96):** C6 outranks C2/C3/C5 for the same task — a residual-sweep failure
 means the workspace holds unfinished task residue, so fixall/unit reruns would either clean it