elliot-stack 1.0.29 → 1.0.33

package/skills/estack-github-issue-tracker/bin/tracker-tools.cjs

@@ -1,1358 +1,1358 @@
-#!/usr/bin/env node
-
-/**
- * tracker-tools.cjs — Data service layer for estack-github-issue-tracker skill.
- *
- * Commands:
- *   startup --tracker <path>              Auth, parse tracker + config, create temp dir, discover issues
- *   fetch-issues --temp-dir <dir> --issues <json-file>  Parallel gh api fetches per issue
- *   compile-report --temp-dir <dir> --date <YYYY-MM-DD>  Build overview from result files
- *   update-tracker --tracker <path> --temp-dir <dir> --date <YYYY-MM-DD>  Apply changes (incl. Goal)
- *   build-tracker --temp-dir <dir> --template <path> --username <name> --tracker <path>  First-run tracker creation
- *
- * All commands return a `today` field with the current date.
- * Zero dependencies. Deterministic output. The agent calls this; it cannot "forget".
- */
-
-'use strict';
-
-const fs = require('fs');
-const path = require('path');
-const os = require('os');
-const { exec, execSync } = require('child_process');
-
-function localDate() {
-  const now = new Date();
-  return `${now.getFullYear()}-${String(now.getMonth() + 1).padStart(2, '0')}-${String(now.getDate()).padStart(2, '0')}`;
-}
-
-function todayTag() {
-  return `today's date is **${localDate()}**, ignore earlier dates`;
-}
-
-// ─── CLI Router ─────────────────────────────────────────────────────────────
-
-const [,, command, ...rawArgs] = process.argv;
-
-const commands = {
-  'compile-report': cmdCompileReport,
-  'update-tracker': cmdUpdateTracker,
-  'startup': cmdStartup,
-  'fetch-issues': cmdFetchIssues,
-  'build-tracker': cmdBuildTracker,
-};
-
-if (!command || !commands[command]) {
-  console.error(`Usage: tracker-tools.cjs <command> [options]`);
-  console.error(`Commands: ${Object.keys(commands).join(', ')}`);
-  process.exit(1);
-}
-
-// Parse --flag value pairs from rawArgs
-function parseFlags(args) {
-  const flags = {};
-  for (let i = 0; i < args.length; i++) {
-    if (args[i].startsWith('--')) {
-      const key = args[i].slice(2);
-      const val = args[i + 1] && !args[i + 1].startsWith('--') ? args[++i] : true;
-      flags[key] = val;
-    } else if (!flags._positional) {
-      flags._positional = args[i];
-    }
-  }
-  return flags;
-}
-
-const flags = parseFlags(rawArgs);
-
-(async () => {
-  try {
-    await commands[command](flags);
-  } catch (err) {
-    console.error(`Error in ${command}: ${err.message}`);
-    process.exit(1);
-  }
-})();
-
-// ─── Frontmatter Parser ────────────────────────────────────────────────────
-
-function parseFrontmatter(content) {
-  const match = content.match(/^---\r?\n([\s\S]*?)\r?\n---/);
-  if (!match) return { meta: {}, body: content };
-
-  const meta = {};
-  const lines = match[1].split(/\r?\n/);
-  for (const line of lines) {
-    const idx = line.indexOf(':');
-    if (idx === -1) continue;
-    const key = line.slice(0, idx).trim();
-    let val = line.slice(idx + 1).trim();
-
-    // Boolean
-    if (val === 'true') val = true;
-    else if (val === 'false') val = false;
-    // Number
-    else if (/^\d+$/.test(val)) val = parseInt(val, 10);
-    // Comma-separated list
-    else if (val.includes(',') && !val.startsWith('"')) {
-      val = val.split(',').map(s => s.trim()).filter(Boolean);
-    }
-    // Strip surrounding quotes
-    else if (val.startsWith('"') && val.endsWith('"')) {
-      val = val.slice(1, -1);
-    }
-
-    meta[key] = val;
-  }
-
-  const body = content.slice(match[0].length).trim();
-  return { meta, body };
-}
-
-// ─── Tracker Parser ─────────────────────────────────────────────────────────
-
-function parseTrackerFile(content) {
-  const result = {
-    username: null,
-    config: null,
-    active_issues: [],
-    closed_issues: [],
-    raw: content,
-  };
-
-  // Extract username
-  const userMatch = content.match(/GitHub username:\s*\*\*(.+?)\*\*/);
-  if (userMatch) result.username = userMatch[1];
-
-  // Extract Config section (between ## Config and next --- or ##)
-  const configMatch = content.match(/## Config\s*\n([\s\S]*?)(?=\n---|\n## (?!Config))/);
-  if (configMatch) {
-    // Strip HTML comments and trim
-    result.config = configMatch[1].replace(/<!--[\s\S]*?-->/g, '').trim() || null;
-  }
-
-  // Split into Active and Closed sections
-  const activeSectionMatch = content.match(
-    /## Active Issues[^\n]*\n([\s\S]*?)(?=\n## (?!Active)|$)/
-  );
-  const closedSectionMatch = content.match(
-    /## Closed[^\n]*\n([\s\S]*?)(?=\n## (?!Closed)|$)/
-  );
-
-  if (activeSectionMatch) {
-    result.active_issues = parseIssueSections(activeSectionMatch[1]);
-  }
-  if (closedSectionMatch) {
-    result.closed_issues = parseClosedSections(closedSectionMatch[1]);
-  }
-
-  return result;
-}
-
-function parseIssueSections(sectionContent) {
-  const issues = [];
-  // Split by ### headers
-  const parts = sectionContent.split(/(?=^### )/m).filter(p => p.trim());
-
-  for (const part of parts) {
-    // Skip HTML comments
-    if (part.trim().startsWith('<!--')) continue;
-
-    const issue = parseIssueBlock(part);
-    if (issue) issues.push(issue);
-  }
-  return issues;
-}
-
-function parseIssueBlock(block) {
-  // Parse header: ### owner/repo#NUMBER — Title
-  const headerMatch = block.match(
-    /^### ([^/]+)\/([^#]+)#(\d+)\s*[—–-]\s*(.+)/m
-  );
-  if (!headerMatch) return null;
-
-  const issue = {
-    owner: headerMatch[1].trim(),
-    repo: headerMatch[2].trim(),
-    number: parseInt(headerMatch[3], 10),
-    title: headerMatch[4].trim(),
-    goal: extractField(block, 'Goal'),
-    role: extractField(block, 'Role'),
-    filed: extractField(block, 'Filed'),
-    last_check_date: null,
-    status_summary: extractField(block, 'Status as of'),
-    what_to_check: extractField(block, 'What to check'),
-    related: extractListField(block, 'Related'),
-    upstream: extractUpstream(block),
-    duplicates: extractNestedItems(block, 'Duplicates found'),
-    adjacent: extractNestedItems(block, 'Adjacent issues found'),
-    next_steps: extractField(block, 'Next steps \\(now\\)'),
-    future: extractField(block, 'Future'),
-  };
-
-  // Extract last check date from "Status as of YYYY-MM-DD:"
-  const dateMatch = block.match(/\*\*Status as of (\d{4}-\d{2}-\d{2}):\*\*/);
-  if (dateMatch) issue.last_check_date = dateMatch[1];
-
-  // Extract History field (multi-line: bullet list under **History:**)
-  issue.history = extractHistoryField(block);
-
-  return issue;
-}
-
-function extractField(block, fieldName) {
-  // Special case: "Status as of YYYY-MM-DD:" has the date embedded in the bold
-  if (fieldName === 'Status as of') {
-    const statusRe = /\*\*Status as of \d{4}-\d{2}-\d{2}:\*\*\s*(.+)/i;
-    const statusMatch = block.match(statusRe);
-    return statusMatch ? statusMatch[1].trim() : null;
-  }
-  const re = new RegExp(`\\*\\*${fieldName}(?:\\s+\\S+)?[.:]*\\*\\*\\s*(.+)`, 'i');
-  const match = block.match(re);
-  return match ? match[1].trim() : null;
-}
-
-function extractListField(block, fieldName) {
-  const val = extractField(block, fieldName);
-  if (!val) return [];
-  return val.split(',').map(s => s.trim()).filter(Boolean);
-}
-
-function extractUpstream(block) {
-  const val = extractField(block, 'Upstream');
-  if (!val || val.toLowerCase() === 'n/a' || val === 'none') return null;
-  const match = val.match(/([^/]+)\/([^#]+)#(\d+)/);
-  if (!match) return null;
-  return { owner: match[1], repo: match[2], number: parseInt(match[3], 10) };
-}
-
-function extractNestedItems(block, sectionName) {
-  const items = [];
-  const re = new RegExp(`\\*\\*${sectionName}[^*]*\\*\\*[:\\s]*\\n([\\s\\S]*?)(?=\\n- \\*\\*[A-Z]|$)`, 'i');
-  const match = block.match(re);
-  if (!match) return items;
-
-  const lines = match[1].split(/\r?\n/);
-  for (const line of lines) {
-    const itemMatch = line.match(/^\s+-\s+\*\*#?(\d+)\*\*\s*[—–-]\s*(.+)/);
-    if (itemMatch) {
-      items.push({
-        number: parseInt(itemMatch[1], 10),
-        detail: itemMatch[2].trim(),
-      });
-    }
-  }
-  return items;
-}
-
-/**
- * Extract history entries from an issue block.
- * Matches all "  - **YYYY-MM-DD:** ..." lines under "**History:**".
- * Returns array of strings like "**2026-03-15:** Filed issue".
- */
-function extractHistoryField(block) {
-  // Find **History:** and collect indented bullet lines until next top-level field or section
-  const historyStart = block.search(/- \*\*History:\*\*/);
-  if (historyStart === -1) return [];
-
-  const afterHistory = block.slice(historyStart);
-  // Collect lines after the **History:** line that match date bullets
-  const lines = afterHistory.split(/\r?\n/);
-  const entries = [];
-  let started = false;
-  for (const line of lines) {
-    if (!started) {
-      if (/- \*\*History:\*\*/.test(line)) { started = true; }
-      continue;
-    }
-    // Match indented date bullets: "  - **YYYY-MM-DD:** ..."
-    const bullet = line.match(/^\s+-\s+(\*\*\d{4}-\d{2}-\d{2}:\*\*.+)/);
-    if (bullet) {
-      entries.push(bullet[1].trim());
-    } else if (line.match(/^- \*\*[A-Z]/) || line.match(/^### /)) {
-      // Next top-level field or section header — stop
-      break;
-    }
-  }
-  return entries;
-}
-
-function parseClosedSections(sectionContent) {
-  const issues = [];
-  const parts = sectionContent.split(/(?=^### )/m).filter(p => p.trim());
-
-  for (const part of parts) {
-    if (part.trim().startsWith('<!--')) continue;
-    const headerMatch = part.match(
-      /^### ([^/]+)\/([^#]+)#(\d+)\s*[—–-]\s*(.+)/m
-    );
-    if (!headerMatch) continue;
-
-    const lines = part.split(/\r?\n/).slice(1).filter(l => l.trim().startsWith('-'));
-    issues.push({
-      owner: headerMatch[1].trim(),
-      repo: headerMatch[2].trim(),
-      number: parseInt(headerMatch[3], 10),
-      title: headerMatch[4].trim(),
-      resolution: lines.map(l => l.replace(/^-\s*/, '').trim()).join(' '),
-    });
-  }
-  return issues;
-}
-
-// ─── Report Compiler ────────────────────────────────────────────────────────
-
-function compileOverviewReport(tempDir, date) {
-  const files = fs.readdirSync(tempDir).filter(f => f.endsWith('.md'));
-  if (files.length === 0) {
-    return { error: 'No result files found in temp directory' };
-  }
-
-  const issueResults = [];
-
-  for (const file of files) {
-    const content = fs.readFileSync(path.join(tempDir, file), 'utf8');
-    const { meta, body } = parseFrontmatter(content);
-
-    if (meta.type === 'issue') {
-      issueResults.push({ meta, body, file });
-    }
-  }
-
-  // Sort: activity first, then by update recency
-  const withActivity = issueResults.filter(r => r.meta.has_activity === true);
-  const noActivity = issueResults.filter(r => r.meta.has_activity !== true);
-
-  // Count repos
-  const repos = new Set(issueResults.map(r => `${r.meta.owner}/${r.meta.repo}`));
-
-  // Build report
-  let report = '';
-
-  report += `## GitHub Issues Check-In — ${date}\n\n`;
-  report += `**Tracking ${issueResults.length} active issues across ${repos.size} repos**`;
-
-  // Find oldest check date for "last check"
-  const checkDates = issueResults
-    .map(r => r.meta.last_check_date)
-    .filter(Boolean)
-    .sort();
-  if (checkDates.length > 0) {
-    report += ` | Last check: ${checkDates[0]}`;
-  }
-  report += '\n\n---\n\n';
-
-  // ── Issues with Activity ──
-  if (withActivity.length > 0) {
-    report += '### Issues with Activity\n\n';
-    report += `These ${withActivity.length} issue${withActivity.length === 1 ? ' has' : 's have'} new comments or state changes since your last check.\n\n`;
-    for (const r of withActivity) {
-      report += buildIssueDetailBlock(r);
-    }
-  }
-
-  // ── No Activity ──
-  if (noActivity.length > 0) {
-    report += '### No Activity\n\n';
-    for (const r of noActivity) {
-      report += buildQuietIssueBlock(r);
-    }
-  }
-
-  // ── Upstream Status ──
-  const upstreamResults = issueResults.filter(r => {
-    const upstreamSection = extractSection(r.body, 'Upstream');
-    return upstreamSection && upstreamSection.trim() !== 'N/A' && upstreamSection.trim() !== 'None';
-  });
-  if (upstreamResults.length > 0) {
-    report += '### Upstream Status\n\n';
-    report += '| Upstream issue | State | Impact |\n';
-    report += '|----------------|-------|--------|\n';
-    for (const r of upstreamResults) {
-      const section = extractSection(r.body, 'Upstream');
-      report += `| ${r.meta.owner}/${r.meta.repo}#${r.meta.number} | ${section.trim()} |\n`;
-    }
-    report += '\n---\n\n';
-  }
-
-  // ── Summary ──
-  const actionItems = withActivity.reduce((count, r) => {
-    const steps = extractSection(r.body, 'Next Steps');
-    if (steps) count += (steps.match(/^- \[/gm) || []).length;
-    return count;
-  }, 0);
-
-  report += '### Summary\n\n';
-  report += `**Action items:** ${actionItems}\n`;
-  if (withActivity.length > 0) {
-    const needAttention = withActivity.slice(0, 3).map(
-      r => `${r.meta.owner}/${r.meta.repo}#${r.meta.number}`
-    );
-    report += `**Issues needing attention:** ${needAttention.join(', ')}\n`;
-  }
-  if (noActivity.length > 0) {
-    report += `**All quiet:** ${noActivity.length} issues with no activity\n`;
-  }
-
-  return { report, issue_count: issueResults.length, activity_count: withActivity.length };
-}
-
-function buildIssueDetailBlock(r) {
-  let block = '';
-  block += `#### ${r.meta.owner}/${r.meta.repo}#${r.meta.number} — ${r.meta.title}\n`;
-  block += `| Field | Value |\n`;
-  block += `|-------|-------|\n`;
-  block += `| **State** | ${r.meta.state || 'Open'}${r.meta.state_changed ? ' [changed]' : ''} |\n`;
-  if (r.meta.labels) {
-    const labels = Array.isArray(r.meta.labels) ? r.meta.labels.join(', ') : r.meta.labels;
-    block += `| **Labels** | ${labels} |\n`;
-  }
-  block += `| **Your role** | ${r.meta.role || 'Unknown'} |\n\n`;
-
-  // Plain English summary of where this issue stands
-  const summary = extractSection(r.body, 'Status Summary');
-  if (summary) {
-    block += `${summary.trim()}\n\n`;
-  }
-
-  // Activity section — what changed since last check
-  const activity = extractSection(r.body, 'Activity');
-  if (activity) {
-    block += `**What changed:**\n${activity}\n\n`;
-  }
-
-  // Duplicates
-  const dupes = extractSection(r.body, 'Known');
-  const newFinds = extractSection(r.body, 'New finds');
-  const hasDupes = dupes && !/^(none|no changes)\.?$/i.test(dupes.trim());
-  const hasNewFinds = newFinds && !/^(none|none found|skipped)\.?/i.test(newFinds.trim());
-  if (hasDupes || hasNewFinds) {
-    block += `**Duplicates & related:**\n`;
-    if (hasDupes) block += `- Known dupes — ${dupes.trim()}\n`;
-    if (hasNewFinds) block += `- New finds — ${newFinds.trim()}\n`;
-    block += '\n';
-  }
-
-  // What you need to do next
-  const steps = extractSection(r.body, 'Next Steps');
-  if (steps) {
-    block += `**What to do next:**\n${steps}\n\n`;
-  }
-
-  // Watch For
-  const watch = extractSection(r.body, 'Watch For');
-  if (watch) {
-    block += `**Watch for:**\n${watch}\n\n`;
-  }
-
-  block += '---\n\n';
-  return block;
-}
-
-function buildQuietIssueBlock(r) {
-  let block = '';
-  block += `#### ${r.meta.owner}/${r.meta.repo}#${r.meta.number} — ${r.meta.title}\n`;
-  const lastDate = r.meta.last_comment_date || r.meta.last_check_date || 'unknown';
-  block += `- Last activity: ${lastDate}\n`;
-
-  // Keep quiet issues compact to avoid report bloat.
-  const summary = extractSection(r.body, 'Status Summary');
-  if (summary) {
-    block += `- Status: ${firstLine(summary)}\n`;
-  }
-
-  const steps = extractSection(r.body, 'Next Steps');
-  if (steps) {
-    block += `- Next: ${firstLine(steps)}\n`;
-  }
-
-  const watch = extractSection(r.body, 'Watch For');
-  if (watch) {
-    block += `- Watch: ${firstLine(watch)}\n`;
-  }
-
-  block += '\n';
-  return block;
-}
-
-function extractSection(body, headerPattern) {
-  // Match ## Header or ### Header, capture until next ## or end
-  const re = new RegExp(
-    `^#{2,4}\\s+(?:[^\\n]*?${headerPattern}[^\\n]*)\\n([\\s\\S]*?)(?=^#{2,4}\\s|$)`,
-    'mi'
-  );
-  const match = body.match(re);
-  return match ? match[1].trim() : null;
-}
-
-function firstLine(text) {
-  if (!text) return '';
-  const line = text.split(/\r?\n/).find(l => l.trim().length > 0) || '';
-  return line.replace(/^\s*[-*]\s*/, '').trim();
-}
-
-// ─── Tracker Updater ────────────────────────────────────────────────────────
-
-function applyTrackerUpdates(trackerContent, tempDir, date) {
-  const files = fs.readdirSync(tempDir).filter(f => f.endsWith('.md'));
-  let updated = trackerContent;
-  const changes = [];
-
-  for (const file of files) {
-    const content = fs.readFileSync(path.join(tempDir, file), 'utf8');
-    const { meta, body } = parseFrontmatter(content);
-
-    if (meta.type !== 'issue') continue;
-
-    const issueKey = `${meta.owner}/${meta.repo}#${meta.number}`;
-
-    // Find this issue's section in the tracker
-    const sectionRe = new RegExp(
-      `(### ${escapeRegex(meta.owner)}/${escapeRegex(meta.repo)}#${meta.number}\\s*[—–-][^\\n]*\\n[\\s\\S]*?)(?=\\n### |\\n## |$)`,
-      'm'
-    );
-    const sectionMatch = updated.match(sectionRe);
-    if (!sectionMatch) continue;
-
-    let section = sectionMatch[1];
-    const originalSection = section;
-
-    // Update "Status as of" date and summary
-    const trackerUpdates = extractSection(body, 'Tracker Updates');
-    if (trackerUpdates) {
-      const statusLine = trackerUpdates.match(/^status_summary:\s*(.+)/m);
-      if (statusLine) {
-        section = section.replace(
-          /\*\*Status as of \d{4}-\d{2}-\d{2}:\*\*\s*.+/,
-          `**Status as of ${date}:** ${statusLine[1].trim()}`
-        );
-      }
-
-      const watchLine = trackerUpdates.match(/^what_to_check:\s*(.+)/m);
-      if (watchLine) {
-        section = section.replace(
-          /\*\*What to check\*\*:\s*.+/,
-          `**What to check:** ${watchLine[1].trim()}`
-        );
-      }
-
-      // Update or add Goal
-      const goalLine = trackerUpdates.match(/^goal:\s*(.+)/m);
-      if (goalLine) {
-        if (section.includes('**Goal:**')) {
-          section = section.replace(
-            /\*\*Goal:\*\*\s*.+/,
-            `**Goal:** ${goalLine[1].trim()}`
-          );
-        } else {
-          // Insert Goal as first field after the header line
-          section = section.replace(
-            /(### [^\n]+\n)/,
-            `$1- **Goal:** ${goalLine[1].trim()}\n`
-          );
-        }
-        changes.push(`Set Goal for ${issueKey}`);
-      }
-    }
-
-    // Handle state change: open → closed (move to Closed section)
-    if (meta.state === 'closed' && meta.state_changed) {
-      // Preserve key fields from the original section before removing it
-      const closedLines = [];
-      closedLines.push(`### ${issueKey} — ${meta.title}`);
-      closedLines.push(`- Closed as of ${date}. ${meta.close_reason || ''}`);
-
-      // Preserve Goal
-      const goalMatch = originalSection.match(/- \*\*Goal:\*\*\s*(.+)/);
-      if (goalMatch) closedLines.push(`- **Goal:** ${goalMatch[1].trim()}`);
-
-      // Preserve Role
-      const roleMatch = originalSection.match(/- \*\*Role:\*\*\s*(.+)/);
-      if (roleMatch) closedLines.push(`- **Role:** ${roleMatch[1].trim()}`);
-
-      // Preserve History section
-      const historyIdx = originalSection.indexOf('- **History:**');
-      if (historyIdx !== -1) {
-        const historyBlock = originalSection.slice(historyIdx);
-        const historyLines = historyBlock.split(/\r?\n/);
-        const preserved = [historyLines[0]]; // "- **History:**"
-        for (let hi = 1; hi < historyLines.length; hi++) {
-          if (/^\s+-\s+\*\*\d{4}-\d{2}-\d{2}:\*\*/.test(historyLines[hi])) {
-            preserved.push(historyLines[hi]);
-          } else if (/^- \*\*[A-Z]/.test(historyLines[hi]) || /^### /.test(historyLines[hi])) {
-            break;
-          }
-        }
-        // Add closing history entry
-        preserved.push(`  - **${date}:** Closed. ${meta.close_reason || ''}`);
-        closedLines.push(...preserved);
-      }
-
-      const closedEntry = closedLines.join('\n') + '\n\n';
-
-      // Remove from active
-      updated = updated.replace(originalSection, '');
-      // Add to closed
-      updated = updated.replace(
-        /(## Closed[^\n]*\n)/,
-        `$1\n${closedEntry}`
-      );
-      changes.push(`Moved ${issueKey} to Closed`);
-      continue;
-    }
-
-    // Add new duplicates
-    if (trackerUpdates) {
-      const newDupeLines = [];
-      const dupeRe = /new_duplicate:\s*#?(\d+)\s*[—–-]\s*(.+)/gm;
-      let dupeMatch;
-      while ((dupeMatch = dupeRe.exec(trackerUpdates)) !== null) {
-        newDupeLines.push(`  - **#${dupeMatch[1]}** — ${dupeMatch[2].trim()}`);
-      }
-      if (newDupeLines.length > 0) {
-        const dupeHeader = `**Duplicates found (${date}):**`;
-        if (section.includes('Duplicates found')) {
-          // Append to existing duplicates section
-          section = section.replace(
-            /(Duplicates found[^*]*\*\*:?\s*\n(?:\s+-[^\n]*\n)*)/,
-            `$1${dupeHeader}\n${newDupeLines.join('\n')}\n`
-          );
-        } else {
-          // Add before Next steps
-          section = section.replace(
-            /(\*\*Next steps)/,
-            `- ${dupeHeader}\n${newDupeLines.join('\n')}\n- $1`
-          );
-        }
-        changes.push(`Added ${newDupeLines.length} duplicates to ${issueKey}`);
-      }
-    }
-
-    // Append history entries from result file
-    if (trackerUpdates) {
-      const newHistoryEntries = [];
-      const historyRe = /^history_entry:\s*(\d{4}-\d{2}-\d{2})\s*\|\s*(.+)/gm;
-      let hm;
-      while ((hm = historyRe.exec(trackerUpdates)) !== null) {
-        newHistoryEntries.push({ date: hm[1], desc: hm[2].trim() });
-      }
-
-      if (newHistoryEntries.length > 0) {
-        const historyHeaderIdx = section.indexOf('- **History:**');
-        if (historyHeaderIdx !== -1) {
-          // History section exists — collect existing entries for dedup
-          const afterHeader = section.slice(historyHeaderIdx);
-          const existingLines = afterHeader.split(/\r?\n/);
-          const existingTexts = new Set();
-          for (const line of existingLines.slice(1)) {
-            const m = line.match(/^\s+-\s+\*\*(\d{4}-\d{2}-\d{2}):\*\*\s*(.+)/);
-            if (m) existingTexts.add(`${m[1]}|${m[2].trim()}`);
-            else if (line.match(/^- \*\*[A-Z]/) || line.match(/^### /)) break;
-          }
-
-          // Build lines to append (deduped)
-          const toAppend = newHistoryEntries
-            .filter(e => !existingTexts.has(`${e.date}|${e.desc}`))
-            .map(e => `  - **${e.date}:** ${e.desc}`);
-
-          if (toAppend.length > 0) {
-            // Find the last history bullet line position and insert after it
-            // We'll insert the new lines before the next top-level field after **History:**
-            const historyBlock = section.slice(historyHeaderIdx);
-            const historyLines = historyBlock.split(/\r?\n/);
-            let lastBulletLine = 0;
-            for (let li = 1; li < historyLines.length; li++) {
-              if (/^\s+-\s+\*\*\d{4}-\d{2}-\d{2}:\*\*/.test(historyLines[li])) {
-                lastBulletLine = li;
-              } else if (/^- \*\*[A-Z]/.test(historyLines[li]) || /^### /.test(historyLines[li])) {
-                break;
-              }
-            }
-            // Insert toAppend after lastBulletLine
-            historyLines.splice(lastBulletLine + 1, 0, ...toAppend);
-            const newHistoryBlock = historyLines.join('\n');
-            section = section.slice(0, historyHeaderIdx) + newHistoryBlock;
-            changes.push(`Appended ${toAppend.length} history entries to ${issueKey}`);
-          }
-        } else {
-          // No history section yet — insert before trailing newline / end of section
-          const insertPoint = section.lastIndexOf('\n');
-          const historyLines = ['- **History:**'];
-          for (const e of newHistoryEntries) {
-            historyLines.push(`  - **${e.date}:** ${e.desc}`);
-          }
-          section = section.slice(0, insertPoint) + '\n' + historyLines.join('\n') + section.slice(insertPoint);
-          changes.push(`Added History section with ${newHistoryEntries.length} entries to ${issueKey}`);
-        }
-      }
-    }
-
-    if (section !== originalSection) {
-      updated = updated.replace(originalSection, section);
-      changes.push(`Updated ${issueKey}`);
-    }
-  }
-
-  return { content: updated, changes };
-}
-
-function escapeRegex(str) {
-  return str.replace(/[.*+?^${}()|[\]\\]/g, '\\$&');
-}
-
-// ─── Command Implementations ────────────────────────────────────────────────
-
-function cmdCompileReport(flags) {
-  const tempDir = flags['temp-dir'];
-  const date = flags.date;
-
-  if (!tempDir || !date) {
-    console.error('Usage: compile-report --temp-dir <dir> --date <YYYY-MM-DD>');
-    process.exit(1);
-  }
-
-  if (!fs.existsSync(tempDir)) {
-    console.error(`Temp directory not found: ${tempDir}`);
-    process.exit(1);
-  }
-
-  const result = compileOverviewReport(tempDir, date);
-  if (result.error) {
-    console.error(result.error);
-    process.exit(1);
-  }
-
-  // Write report to temp dir as well for validation
-  fs.writeFileSync(path.join(tempDir, '_compiled-report.md'), result.report, 'utf8');
-
-  // Output the report to stdout
-  console.log(result.report);
-
-  // Write metadata to stderr for the agent to parse
-  console.error(JSON.stringify({
-    today: todayTag(),
-    issue_count: result.issue_count,
-    activity_count: result.activity_count,
-    report_file: path.join(tempDir, '_compiled-report.md'),
-  }));
-}
-
-function cmdUpdateTracker(flags) {
-  const trackerPath = flags.tracker;
-  const tempDir = flags['temp-dir'];
-  const date = flags.date;
-
-  if (!trackerPath || !tempDir || !date) {
-    console.error('Usage: update-tracker --tracker <path> --temp-dir <dir> --date <YYYY-MM-DD>');
-    process.exit(1);
-  }
-
-  if (!fs.existsSync(trackerPath)) {
-    console.error(`Tracker file not found: ${trackerPath}`);
-    process.exit(1);
-  }
-
-  const content = fs.readFileSync(trackerPath, 'utf8');
-  const { content: updated, changes } = applyTrackerUpdates(content, tempDir, date);
-
-  if (changes.length === 0) {
-    console.log(JSON.stringify({ today: todayTag(), updated: false, changes: [] }));
-    return;
-  }
-
-  fs.writeFileSync(trackerPath, updated, 'utf8');
-  console.log(JSON.stringify({ today: todayTag(), updated: true, changes }));
-}
-
-// ─── Async Helper ──────────────────────────────────────────────────────────
-
-function execAsync(cmd) {
-  return new Promise((resolve, reject) => {
-    exec(cmd, { maxBuffer: 10 * 1024 * 1024 }, (err, stdout, stderr) => {
-      if (err) reject(err);
-      else resolve(stdout);
-    });
-  });
-}
-
-/**
- * Run promises in batches to avoid rate limits.
- * @param {Array<() => Promise>} tasks - Array of functions returning promises
- * @param {number} concurrency - Max concurrent tasks
- * @returns {Promise<Array>} Results in same order as tasks
- */
-async function batchRun(tasks, concurrency = 15) {
-  const results = [];
-  for (let i = 0; i < tasks.length; i += concurrency) {
-    const batch = tasks.slice(i, i + concurrency);
-    const batchResults = await Promise.all(batch.map(fn => fn()));
-    results.push(...batchResults);
-  }
-  return results;
-}
-
-// ─── Command: startup ──────────────────────────────────────────────────────
-
-async function cmdStartup(flags) {
-  const trackerPath = flags.tracker;
-  if (!trackerPath) {
-    console.error('Usage: startup --tracker <path>');
-    process.exit(1);
-  }
-
-  // Step 1: Check gh auth
-  let username = null;
-  try {
-    const authOutput = execSync('gh auth status 2>&1', { encoding: 'utf8' });
-    const userMatch = authOutput.match(/Logged in to github\.com.*account\s+(\S+)/i)
-      || authOutput.match(/Logged in to github\.com\s+as\s+(\S+)/i)
-      || authOutput.match(/account\s+(\S+)/i);
-    if (userMatch) username = userMatch[1];
-  } catch (err) {
-    console.log(JSON.stringify({
-      script_ok: true, auth: false, today: todayTag(),
-      error: 'Not authenticated with GitHub. Run `gh auth login` in your terminal to fix this.'
-    }));
-    return;
-  }
-
-  if (!username) {
-    // Try alternate extraction from gh api
-    try {
-      username = execSync('gh api user --jq .login', { encoding: 'utf8' }).trim();
-    } catch (_) {
-      console.log(JSON.stringify({
-        script_ok: true, auth: false, today: todayTag(),
-        error: 'Not authenticated with GitHub. Run `gh auth login` in your terminal to fix this.'
-      }));
-      return;
-    }
-  }
-
-  // Step 2: Parse tracker (or set empty defaults if no tracker)
-  const thirtyDaysAgo = new Date(Date.now() - 30 * 24 * 60 * 60 * 1000)
-    .toISOString().split('T')[0];
-
-  let trackerExists = false;
-  let parsed = { username: null, active_issues: [], closed_issues: [], raw: '' };
-
-  if (fs.existsSync(trackerPath)) {
-    const content = fs.readFileSync(trackerPath, 'utf8');
-    if (content.trim()) {
-      trackerExists = true;
-      parsed = parseTrackerFile(content);
-      // Default null last_check_date to 30 days ago
-      for (const issue of parsed.active_issues) {
-        if (!issue.last_check_date) issue.last_check_date = thirtyDaysAgo;
-      }
-    }
-  }
-
-  const oldestCheckDate = parsed.active_issues
-    .map(i => i.last_check_date)
-    .filter(Boolean)
-    .sort()[0] || thirtyDaysAgo;
-
-  const allTrackedNumbers = [
-    ...parsed.active_issues.map(i => `${i.owner}/${i.repo}#${i.number}`),
-    ...parsed.closed_issues.map(i => `${i.owner}/${i.repo}#${i.number}`),
-  ];
-
-  // Step 3: Run two gh api search queries in parallel (always, even without tracker)
-  const searchDate = oldestCheckDate;
-  const openQuery = `search/issues?q=involves:${username}+is:open+updated:>${searchDate}&per_page=100`;
-  const closedQuery = `search/issues?q=involves:${username}+is:closed+closed:>${thirtyDaysAgo}&per_page=50`;
-
-  let openResults = [];
-  let closedResults = [];
-  let searchErrors = [];
-
-  // Run searches individually so one failure doesn't block the other
-  try {
-    const openRaw = await execAsync(`gh api "${openQuery}" --jq ".items"`);
-    openResults = JSON.parse(openRaw || '[]');
-  } catch (err) {
-    searchErrors.push(`open search failed: ${err.message}`);
-    openResults = [];
-  }
-
-  try {
-    const closedRaw = await execAsync(`gh api "${closedQuery}" --jq ".items"`);
-    closedResults = JSON.parse(closedRaw || '[]');
-  } catch (err) {
-    searchErrors.push(`closed search failed: ${err.message}`);
-    closedResults = [];
-  }
-
-  // Step 4: Identify new, reopened, recently closed
-  function issueKey(item) {
-    const urlParts = (item.repository_url || '').split('/');
-    const owner = urlParts[urlParts.length - 2];
-    const repo = urlParts[urlParts.length - 1];
-    return `${owner}/${repo}#${item.number}`;
-  }
-
-  function issueInfo(item) {
-    const urlParts = (item.repository_url || '').split('/');
-    return {
-      owner: urlParts[urlParts.length - 2],
-      repo: urlParts[urlParts.length - 1],
-      number: item.number,
-      title: item.title,
-      state: item.state,
-      updated_at: item.updated_at,
-      created_at: item.created_at,
-      html_url: item.html_url,
-      labels: (item.labels || []).map(l => l.name),
-      user: item.user ? item.user.login : null,
-    };
-  }
-
-  const trackedSet = new Set(allTrackedNumbers);
-  const closedNumbers = new Set(
-    parsed.closed_issues.map(i => `${i.owner}/${i.repo}#${i.number}`)
-  );
-
-  const newIssues = openResults
-    .filter(item => !trackedSet.has(issueKey(item)))
-    .map(issueInfo);
-
-  const reopenedIssues = openResults
-    .filter(item => closedNumbers.has(issueKey(item)))
-    .map(issueInfo);
-
-  const recentlyClosed = closedResults.map(issueInfo);
-
-  // Create temp directory for this session
-  const prefix = 'giu-checkin-';
-  const tempDir = fs.mkdtempSync(path.join(os.tmpdir(), prefix));
-
-  console.log(JSON.stringify({
-    script_ok: true,
-    auth: true,
-    today: todayTag(),
-    username,
-    temp_dir: tempDir,
-    tracker_exists: trackerExists,
-    tracker_path: trackerPath,
-    config: parsed.config,
-    tracker_data: {
-      username: parsed.username,
-      active_issues: parsed.active_issues,
-      closed_issues: parsed.closed_issues,
-      all_tracked_numbers: allTrackedNumbers,
-      oldest_check_date: oldestCheckDate,
-      raw: parsed.raw,
-    },
-    new_issues: newIssues,
-    reopened_issues: reopenedIssues,
-    recently_closed: recentlyClosed,
-    search_errors: searchErrors,
-  }, null, 2));
-}
-
-// ─── Command: fetch-issues ─────────────────────────────────────────────────
-
-async function cmdFetchIssues(flags) {
-  const tempDir = flags['temp-dir'];
-  const issuesFile = flags.issues;
-
-  if (!tempDir || !issuesFile) {
-    console.error('Usage: fetch-issues --temp-dir <dir> --issues <json-file>');
-    process.exit(1);
-  }
-
-  if (!fs.existsSync(issuesFile)) {
-    console.error(`Issues file not found: ${issuesFile}`);
-    process.exit(1);
-  }
-
-  const issues = JSON.parse(fs.readFileSync(issuesFile, 'utf8'));
-  const errors = [];
-  const files = [];
-
-  // Build all fetch tasks
-  const allTasks = [];
-  const taskMap = []; // Maps task index to { issueIdx, type }
-
-  // Fetch raw JSON from gh api (no --jq — process in Node to avoid Windows quoting issues)
-  function ghFetch(endpoint) {
-    return execAsync(`gh api ${endpoint}`)
-      .then(raw => JSON.parse(raw || '{}'))
-      .catch(e => ({ _error: e.message }));
-  }
-
-  for (let i = 0; i < issues.length; i++) {
-    const iss = issues[i];
-    const { owner, repo, number, last_check_date, known_dupes, upstream } = iss;
-    const issueEndpoint = `repos/${owner}/${repo}/issues/${number}`;
-
-    // Issue data (single call — extract metadata + body + author in Node)
-    allTasks.push(() => ghFetch(issueEndpoint));
-    taskMap.push({ issueIdx: i, type: 'issue' });
-
-    // Comments (all of them — filter by date in Node)
-    allTasks.push(() => execAsync(`gh api ${issueEndpoint}/comments --paginate`)
-      .then(raw => JSON.parse(raw || '[]'))
-      .catch(e => ({ _error: e.message })));
-    taskMap.push({ issueIdx: i, type: 'comments' });
-
-    // Known dupe state checks
-    const dupes = known_dupes || [];
-    for (const dupe of dupes) {
-      const dupeMatch = String(dupe).match(/(?:([^/]+)\/([^#]+))?#?(\d+)/);
-      if (dupeMatch) {
-        const dOwner = dupeMatch[1] || owner;
-        const dRepo = dupeMatch[2] || repo;
-        const dNumber = dupeMatch[3];
-        allTasks.push(() => ghFetch(`repos/${dOwner}/${dRepo}/issues/${dNumber}`));
-        taskMap.push({ issueIdx: i, type: 'dupe', dupeKey: `${dOwner}/${dRepo}#${dNumber}` });
-      }
-    }
-
-    // Upstream check
-    if (upstream) {
-      const uMatch = String(upstream).match(/([^/]+)\/([^#]+)#(\d+)/);
-      if (uMatch) {
-        allTasks.push(() => ghFetch(`repos/${uMatch[1]}/${uMatch[2]}/issues/${uMatch[3]}`));
-        taskMap.push({ issueIdx: i, type: 'upstream' });
-      }
-    }
-  }
-
-  // Run all tasks with batching (max 15 concurrent)
-  const results = await batchRun(allTasks, 15);
-
-  // Organize results by issue — process raw API responses in Node
-  const issueData = issues.map(() => ({
-    metadata: null,
-    body: null,
-    comments: null,
-    dupe_states: {},
-    upstream_state: null,
-    cross_references: [],
-    urls: [],
-  }));
-
-  for (let t = 0; t < results.length; t++) {
-    const { issueIdx, type, dupeKey } = taskMap[t];
-    const raw = results[t];
-
-    switch (type) {
-      case 'issue': {
-        // Extract metadata and body from the single issue response
-        if (raw && !raw._error) {
-          issueData[issueIdx].metadata = {
-            state: raw.state,
-            labels: (raw.labels || []).map(l => l.name),
-            comments: raw.comments,
-            updated: raw.updated_at,
-            created: raw.created_at,
-            html_url: raw.html_url,
-          };
-          issueData[issueIdx].body = {
-            title: raw.title,
-            body: raw.body,
-            author: raw.user ? raw.user.login : null,
-          };
-        } else {
-          issueData[issueIdx].metadata = raw;
-          issueData[issueIdx].body = raw;
-        }
-        break;
-      }
-      case 'comments': {
-        // Filter comments by last_check_date in Node
-        const iss = issues[issueIdx];
-        if (Array.isArray(raw)) {
-          const filtered = iss.last_check_date
-            ? raw.filter(c => c.created_at > iss.last_check_date)
-            : raw;
-          issueData[issueIdx].comments = filtered.map(c => ({
-            author: c.user ? c.user.login : null,
-            date: c.created_at ? c.created_at.split('T')[0] : null,
-            body: c.body,
-          }));
-        } else {
-          issueData[issueIdx].comments = raw;
-        }
-        break;
-      }
-      case 'dupe': {
-        if (raw && !raw._error) {
-          issueData[issueIdx].dupe_states[dupeKey] = {
-            state: raw.state,
-            updated: raw.updated_at,
-          };
-        } else {
-          issueData[issueIdx].dupe_states[dupeKey] = raw;
-        }
-        break;
-      }
-      case 'upstream': {
-        if (raw && !raw._error) {
-          issueData[issueIdx].upstream_state = {
-            state: raw.state,
-            labels: (raw.labels || []).map(l => l.name),
-            updated: raw.updated_at,
-          };
-        } else {
-          issueData[issueIdx].upstream_state = raw;
-        }
-        break;
-      }
-    }
-  }
-
-  // Post-process: extract cross_references and urls from body + comments
-  for (let i = 0; i < issues.length; i++) {
-    const data = issueData[i];
-    const allText = collectText(data);
-
-    // Extract #NUMBER and owner/repo#NUMBER patterns
-    const crossRefs = new Set();
-    const refRe = /(?:([a-zA-Z0-9_.-]+\/[a-zA-Z0-9_.-]+))?#(\d+)/g;
-    let m;
-    while ((m = refRe.exec(allText)) !== null) {
-      crossRefs.add(m[1] ? `${m[1]}#${m[2]}` : `#${m[2]}`);
-    }
-    data.cross_references = [...crossRefs];
-
-    // Extract URLs
-    const urlSet = new Set();
-    const urlRe = /https?:\/\/[^\s)>\]"']+/g;
-    while ((m = urlRe.exec(allText)) !== null) {
-      urlSet.add(m[0].replace(/[.,;:]+$/, ''));
-    }
-    data.urls = [...urlSet];
-
-    // Write raw JSON per issue
-    const iss = issues[i];
-    const outFile = path.join(tempDir, `raw-${iss.owner}-${iss.repo}-${iss.number}.json`);
-    try {
-      fs.writeFileSync(outFile, JSON.stringify(data, null, 2), 'utf8');
-      files.push(outFile);
-    } catch (err) {
-      errors.push(`Failed to write ${outFile}: ${err.message}`);
-    }
-  }
-
-  console.log(JSON.stringify({ today: todayTag(), fetched: files.length, files, errors }));
-}
-
-/**
- * Collect all text content from issue data for cross-reference/URL extraction.
- */
-function collectText(data) {
-  const parts = [];
-  if (data.body) {
-    if (typeof data.body === 'string') parts.push(data.body);
-    else if (data.body.body) parts.push(data.body.body);
-  }
-  if (data.comments) {
-    if (typeof data.comments === 'string') parts.push(data.comments);
-    else if (Array.isArray(data.comments)) {
-      for (const c of data.comments) {
-        if (typeof c === 'string') parts.push(c);
-        else if (c && c.body) parts.push(c.body);
-      }
-    }
-  }
-  return parts.join('\n');
-}
-
-// ─── Command: build-tracker ────────────────────────────────────────────────
-
-function cmdBuildTracker(flags) {
-  const tempDir = flags['temp-dir'];
-  const templatePath = flags.template;
-  const username = flags.username;
-  const trackerPath = flags.tracker;
-  const closedJsonPath = flags['closed-json'];
-  const date = flags.date || localDate();
-
-  if (!tempDir || !templatePath || !username || !trackerPath) {
-    console.error('Usage: build-tracker --temp-dir <dir> --template <path> --username <name> --tracker <path> [--closed-json <path>]');
-    process.exit(1);
-  }
-
-  // Step 1: Read template
-  if (!fs.existsSync(templatePath)) {
-    console.error(`Template not found: ${templatePath}`);
-    process.exit(1);
-  }
-  let tracker = fs.readFileSync(templatePath, 'utf8');
-
-  // Step 2: Replace USERNAME_HERE
-  tracker = tracker.replace(/USERNAME_HERE/g, username);
-
-  // Step 3: Read all issue-*.md result files from temp dir
-  if (!fs.existsSync(tempDir)) {
-    console.error(`Temp directory not found: ${tempDir}`);
-    process.exit(1);
-  }
-
-  const resultFiles = fs.readdirSync(tempDir).filter(f =>
-    f.startsWith('issue-') && f.endsWith('.md')
-  );
-
-  const entries = [];
-  for (const file of resultFiles) {
-    const content = fs.readFileSync(path.join(tempDir, file), 'utf8');
-    const { meta, body } = parseFrontmatter(content);
-    if (meta.type !== 'issue') continue;
-
-    // Build tracker entry from result file
-    const entry = buildTrackerEntry(meta, body, date);
-    entries.push(entry);
-  }
-
-  // Step 4: Insert active entries into tracker
-  const activeEntriesText = entries.join('\n');
-  tracker = tracker.replace(
-    /(## Active Issues[^\n]*\n)([\s\S]*?)(## Closed)/,
-    `$1\n${activeEntriesText}\n$3`
-  );
-
-  // Step 5: Handle closed issues if provided
-  let closedCount = 0;
-  if (closedJsonPath && fs.existsSync(closedJsonPath)) {
-    const closedIssues = JSON.parse(fs.readFileSync(closedJsonPath, 'utf8'));
-    closedCount = closedIssues.length;
-    let closedText = '';
-    for (const ci of closedIssues) {
-      closedText += `### ${ci.owner}/${ci.repo}#${ci.number} — ${ci.title}\n`;
-      closedText += `- ${ci.resolution || 'Closed.'}\n\n`;
-    }
-    if (closedText) {
-      tracker = tracker.replace(
-        /(## Closed[^\n]*\n)/,
-        `$1\n${closedText}`
-      );
-    }
-  }
-
-  // Step 6: Write tracker
-  fs.writeFileSync(trackerPath, tracker, 'utf8');
-
-  console.log(JSON.stringify({
-    today: todayTag(),
-    written: true,
-    path: trackerPath,
-    active_count: entries.length,
-    closed_count: closedCount,
-  }));
-}
-
-/**
- * Build a tracker entry string from result file frontmatter and body.
- */
-function buildTrackerEntry(meta, body, date) {
-  const lines = [];
-  lines.push(`### ${meta.owner}/${meta.repo}#${meta.number} — ${meta.title}`);
-
-  // Goal — from tracker updates or frontmatter
-  const trackerUpdatesForGoal = extractSection(body, 'Tracker Updates') || '';
-  const goalLine = trackerUpdatesForGoal.match(/^goal:\s*(.+)/m);
-  if (goalLine) {
-    lines.push(`- **Goal:** ${goalLine[1].trim()}`);
-  }
-
-  // Role
-  lines.push(`- **Role:** ${meta.role || 'Unknown'}`);
-
-  // Filed
-  if (meta.filed) {
-    lines.push(`- **Filed:** ${meta.filed}`);
-  }
-
-  // Status — from ## Status Summary + labels
-  const statusSummary = extractSection(body, 'Status Summary') || 'Open.';
-  const labels = meta.labels
-    ? (Array.isArray(meta.labels) ? meta.labels.join(', ') : meta.labels)
-    : '';
-  const trackerUpdates = extractSection(body, 'Tracker Updates') || '';
-  const statusLine = trackerUpdates.match(/^status_summary:\s*(.+)/m);
-  const statusText = statusLine
-    ? statusLine[1].trim()
-    : `${meta.state || 'Open'}. Labels: ${labels}. ${statusSummary.split('\n')[0]}`;
-  const dateStr = date || localDate();
-  lines.push(`- **Status as of ${dateStr}:** ${statusText}`);
-
-  // What to check — from ## Watch For or tracker updates
-  const watchLine = trackerUpdates.match(/^what_to_check:\s*(.+)/m);
-  const watchFor = watchLine
-    ? watchLine[1].trim()
-    : (extractSection(body, 'Watch For') || 'Monitor for updates.').split('\n')[0].replace(/^-\s*/, '');
-  lines.push(`- **What to check:** ${watchFor}`);
-
-  // Related — from ## Cross-References
-  const crossRefs = extractSection(body, 'Cross-References');
-  if (crossRefs && crossRefs.trim() !== 'None' && crossRefs.trim() !== 'None found.') {
-    lines.push(`- **Related:** ${crossRefs.split('\n')[0].trim()}`);
-  }
-
-  // Upstream
-  const upstream = extractSection(body, 'Upstream');
-  if (upstream && upstream.trim() !== 'N/A' && upstream.trim() !== 'None') {
-    lines.push(`- **Upstream:** ${upstream.split('\n')[0].trim()}`);
-  }
-
-  // Key Context -> workaround / future
-  const keyContext = extractSection(body, 'Key Context');
-  if (keyContext && keyContext.trim() !== 'N/A') {
-    // Check for workaround
-    const workaroundMatch = keyContext.match(/[Ww]orkaround:?\s*(.+)/);
-    if (workaroundMatch) {
-      lines.push(`- **Workaround:** ${workaroundMatch[1].trim()}`);
-    }
-  }
-
-  // Duplicates from ## Duplicates and Related
-  const dupeSection = extractSection(body, 'Duplicates and Related')
-    || extractSection(body, 'Known');
-  if (dupeSection) {
-    const dupeLines = dupeSection.split('\n').filter(l => l.match(/^\s*-?\s*[#*]/));
-    if (dupeLines.length > 0) {
-      lines.push(`- **Duplicates found (${dateStr}):**`);
-      for (const dl of dupeLines) {
-        const cleaned = dl.replace(/^[\s-]*/, '  - ');
-        lines.push(cleaned);
-      }
-    }
-  }
-
-  // Next steps
-  const nextSteps = extractSection(body, 'Next Steps');
-  if (nextSteps && nextSteps.trim() !== 'None' && nextSteps.trim() !== 'None — no action needed.') {
-    lines.push(`- **Next steps (now):** ${nextSteps.split('\n')[0].replace(/^-\s*\[.\]\s*/, '').trim()}`);
-  } else {
-    lines.push(`- **Next steps (now):** None — no action needed.`);
-  }
-
-  // Future — from Key Context
-  if (keyContext && keyContext.trim() !== 'N/A') {
-    lines.push(`- **Future:** ${keyContext.split('\n')[0].replace(/^-\s*/, '').trim()}`);
-  }
-
-  // History — from history_entry lines in Tracker Updates
-  const historyEntries = [];
-  if (trackerUpdates) {
-    const historyRe = /^history_entry:\s*(\d{4}-\d{2}-\d{2})\s*\|\s*(.+)/gm;
-    let hm;
-    while ((hm = historyRe.exec(trackerUpdates)) !== null) {
-      historyEntries.push(`  - **${hm[1]}:** ${hm[2].trim()}`);
-    }
-  }
-  if (historyEntries.length === 0) {
-    // Default entry on initial build
-    historyEntries.push(`  - **${dateStr}:** Added to tracker`);
-  }
-  lines.push(`- **History:**`);
-  for (const he of historyEntries) {
-    lines.push(he);
-  }
-
-  lines.push('');
-  return lines.join('\n');
-}
+#!/usr/bin/env node
+
+/**
+ * tracker-tools.cjs — Data service layer for estack-github-issue-tracker skill.
+ *
+ * Commands:
+ *   startup --tracker <path>              Auth, parse tracker + config, create temp dir, discover issues
+ *   fetch-issues --temp-dir <dir> --issues <json-file>  Parallel gh api fetches per issue
+ *   compile-report --temp-dir <dir> --date <YYYY-MM-DD>  Build overview from result files
+ *   update-tracker --tracker <path> --temp-dir <dir> --date <YYYY-MM-DD>  Apply changes (incl. Goal)
+ *   build-tracker --temp-dir <dir> --template <path> --username <name> --tracker <path>  First-run tracker creation
+ *
+ * All commands return a `today` field with the current date.
+ * Zero dependencies. Deterministic output. The agent calls this; it cannot "forget".
+ */
+
+'use strict';
+
+const fs = require('fs');
+const path = require('path');
+const os = require('os');
+const { exec, execSync } = require('child_process');
+
+function localDate() {
+  const now = new Date();
+  return `${now.getFullYear()}-${String(now.getMonth() + 1).padStart(2, '0')}-${String(now.getDate()).padStart(2, '0')}`;
+}
+
+function todayTag() {
+  return `today's date is **${localDate()}**, ignore earlier dates`;
+}
+
+// ─── CLI Router ─────────────────────────────────────────────────────────────
+
+const [,, command, ...rawArgs] = process.argv;
+
+const commands = {
+  'compile-report': cmdCompileReport,
+  'update-tracker': cmdUpdateTracker,
+  'startup': cmdStartup,
+  'fetch-issues': cmdFetchIssues,
+  'build-tracker': cmdBuildTracker,
+};
+
+if (!command || !commands[command]) {
+  console.error(`Usage: tracker-tools.cjs <command> [options]`);
+  console.error(`Commands: ${Object.keys(commands).join(', ')}`);
+  process.exit(1);
+}
+
+// Parse --flag value pairs from rawArgs
+function parseFlags(args) {
+  const flags = {};
+  for (let i = 0; i < args.length; i++) {
+    if (args[i].startsWith('--')) {
+      const key = args[i].slice(2);
+      const val = args[i + 1] && !args[i + 1].startsWith('--') ? args[++i] : true;
+      flags[key] = val;
+    } else if (!flags._positional) {
+      flags._positional = args[i];
+    }
+  }
+  return flags;
+}
+
+const flags = parseFlags(rawArgs);
+
+(async () => {
+  try {
+    await commands[command](flags);
+  } catch (err) {
+    console.error(`Error in ${command}: ${err.message}`);
+    process.exit(1);
+  }
+})();
+
+// ─── Frontmatter Parser ────────────────────────────────────────────────────
+
+function parseFrontmatter(content) {
+  const match = content.match(/^---\r?\n([\s\S]*?)\r?\n---/);
+  if (!match) return { meta: {}, body: content };
+
+  const meta = {};
+  const lines = match[1].split(/\r?\n/);
+  for (const line of lines) {
+    const idx = line.indexOf(':');
+    if (idx === -1) continue;
+    const key = line.slice(0, idx).trim();
+    let val = line.slice(idx + 1).trim();
+
+    // Boolean
+    if (val === 'true') val = true;
+    else if (val === 'false') val = false;
+    // Number
+    else if (/^\d+$/.test(val)) val = parseInt(val, 10);
+    // Comma-separated list
+    else if (val.includes(',') && !val.startsWith('"')) {
+      val = val.split(',').map(s => s.trim()).filter(Boolean);
+    }
+    // Strip surrounding quotes
+    else if (val.startsWith('"') && val.endsWith('"')) {
+      val = val.slice(1, -1);
+    }
+
+    meta[key] = val;
+  }
+
+  const body = content.slice(match[0].length).trim();
+  return { meta, body };
+}
+
+// ─── Tracker Parser ─────────────────────────────────────────────────────────
+
+function parseTrackerFile(content) {
+  const result = {
+    username: null,
+    config: null,
+    active_issues: [],
+    closed_issues: [],
+    raw: content,
+  };
+
+  // Extract username
+  const userMatch = content.match(/GitHub username:\s*\*\*(.+?)\*\*/);
+  if (userMatch) result.username = userMatch[1];
+
+  // Extract Config section (between ## Config and next --- or ##)
+  const configMatch = content.match(/## Config\s*\n([\s\S]*?)(?=\n---|\n## (?!Config))/);
+  if (configMatch) {
+    // Strip HTML comments and trim
+    result.config = configMatch[1].replace(/<!--[\s\S]*?-->/g, '').trim() || null;
+  }
+
+  // Split into Active and Closed sections
+  const activeSectionMatch = content.match(
+    /## Active Issues[^\n]*\n([\s\S]*?)(?=\n## (?!Active)|$)/
+  );
+  const closedSectionMatch = content.match(
+    /## Closed[^\n]*\n([\s\S]*?)(?=\n## (?!Closed)|$)/
+  );
+
+  if (activeSectionMatch) {
+    result.active_issues = parseIssueSections(activeSectionMatch[1]);
+  }
+  if (closedSectionMatch) {
+    result.closed_issues = parseClosedSections(closedSectionMatch[1]);
+  }
+
+  return result;
+}
+
+function parseIssueSections(sectionContent) {
+  const issues = [];
+  // Split by ### headers
+  const parts = sectionContent.split(/(?=^### )/m).filter(p => p.trim());
+
+  for (const part of parts) {
+    // Skip HTML comments
+    if (part.trim().startsWith('<!--')) continue;
+
+    const issue = parseIssueBlock(part);
+    if (issue) issues.push(issue);
+  }
+  return issues;
+}
+
+function parseIssueBlock(block) {
+  // Parse header: ### owner/repo#NUMBER — Title
+  const headerMatch = block.match(
+    /^### ([^/]+)\/([^#]+)#(\d+)\s*[—–-]\s*(.+)/m
+  );
+  if (!headerMatch) return null;
+
+  const issue = {
+    owner: headerMatch[1].trim(),
+    repo: headerMatch[2].trim(),
+    number: parseInt(headerMatch[3], 10),
+    title: headerMatch[4].trim(),
+    goal: extractField(block, 'Goal'),
+    role: extractField(block, 'Role'),
+    filed: extractField(block, 'Filed'),
+    last_check_date: null,
+    status_summary: extractField(block, 'Status as of'),
+    what_to_check: extractField(block, 'What to check'),
+    related: extractListField(block, 'Related'),
+    upstream: extractUpstream(block),
+    duplicates: extractNestedItems(block, 'Duplicates found'),
+    adjacent: extractNestedItems(block, 'Adjacent issues found'),
+    next_steps: extractField(block, 'Next steps \\(now\\)'),
+    future: extractField(block, 'Future'),
+  };
+
+  // Extract last check date from "Status as of YYYY-MM-DD:"
+  const dateMatch = block.match(/\*\*Status as of (\d{4}-\d{2}-\d{2}):\*\*/);
+  if (dateMatch) issue.last_check_date = dateMatch[1];
+
+  // Extract History field (multi-line: bullet list under **History:**)
+  issue.history = extractHistoryField(block);
+
+  return issue;
+}
+
+function extractField(block, fieldName) {
+  // Special case: "Status as of YYYY-MM-DD:" has the date embedded in the bold
+  if (fieldName === 'Status as of') {
+    const statusRe = /\*\*Status as of \d{4}-\d{2}-\d{2}:\*\*\s*(.+)/i;
+    const statusMatch = block.match(statusRe);
+    return statusMatch ? statusMatch[1].trim() : null;
+  }
+  const re = new RegExp(`\\*\\*${fieldName}(?:\\s+\\S+)?[.:]*\\*\\*\\s*(.+)`, 'i');
+  const match = block.match(re);
+  return match ? match[1].trim() : null;
+}
+
+function extractListField(block, fieldName) {
+  const val = extractField(block, fieldName);
+  if (!val) return [];
+  return val.split(',').map(s => s.trim()).filter(Boolean);
+}
+
+function extractUpstream(block) {
+  const val = extractField(block, 'Upstream');
+  if (!val || val.toLowerCase() === 'n/a' || val === 'none') return null;
+  const match = val.match(/([^/]+)\/([^#]+)#(\d+)/);
+  if (!match) return null;
+  return { owner: match[1], repo: match[2], number: parseInt(match[3], 10) };
+}
+
+function extractNestedItems(block, sectionName) {
+  const items = [];
+  const re = new RegExp(`\\*\\*${sectionName}[^*]*\\*\\*[:\\s]*\\n([\\s\\S]*?)(?=\\n- \\*\\*[A-Z]|$)`, 'i');
+  const match = block.match(re);
+  if (!match) return items;
+
+  const lines = match[1].split(/\r?\n/);
+  for (const line of lines) {
+    const itemMatch = line.match(/^\s+-\s+\*\*#?(\d+)\*\*\s*[—–-]\s*(.+)/);
+    if (itemMatch) {
+      items.push({
+        number: parseInt(itemMatch[1], 10),
+        detail: itemMatch[2].trim(),
+      });
+    }
+  }
+  return items;
+}
+
+/**
+ * Extract history entries from an issue block.
+ * Matches all "  - **YYYY-MM-DD:** ..." lines under "**History:**".
+ * Returns array of strings like "**2026-03-15:** Filed issue".
+ */
+function extractHistoryField(block) {
+  // Find **History:** and collect indented bullet lines until next top-level field or section
+  const historyStart = block.search(/- \*\*History:\*\*/);
+  if (historyStart === -1) return [];
+
+  const afterHistory = block.slice(historyStart);
+  // Collect lines after the **History:** line that match date bullets
+  const lines = afterHistory.split(/\r?\n/);
+  const entries = [];
+  let started = false;
+  for (const line of lines) {
+    if (!started) {
+      if (/- \*\*History:\*\*/.test(line)) { started = true; }
+      continue;
+    }
+    // Match indented date bullets: "  - **YYYY-MM-DD:** ..."
+    const bullet = line.match(/^\s+-\s+(\*\*\d{4}-\d{2}-\d{2}:\*\*.+)/);
+    if (bullet) {
+      entries.push(bullet[1].trim());
+    } else if (line.match(/^- \*\*[A-Z]/) || line.match(/^### /)) {
+      // Next top-level field or section header — stop
+      break;
+    }
+  }
+  return entries;
+}
+
+function parseClosedSections(sectionContent) {
+  const issues = [];
+  const parts = sectionContent.split(/(?=^### )/m).filter(p => p.trim());
+
+  for (const part of parts) {
+    if (part.trim().startsWith('<!--')) continue;
+    const headerMatch = part.match(
+      /^### ([^/]+)\/([^#]+)#(\d+)\s*[—–-]\s*(.+)/m
+    );
+    if (!headerMatch) continue;
+
+    const lines = part.split(/\r?\n/).slice(1).filter(l => l.trim().startsWith('-'));
+    issues.push({
+      owner: headerMatch[1].trim(),
+      repo: headerMatch[2].trim(),
+      number: parseInt(headerMatch[3], 10),
+      title: headerMatch[4].trim(),
+      resolution: lines.map(l => l.replace(/^-\s*/, '').trim()).join(' '),
+    });
+  }
+  return issues;
+}
+
+// ─── Report Compiler ────────────────────────────────────────────────────────
+
+function compileOverviewReport(tempDir, date) {
+  const files = fs.readdirSync(tempDir).filter(f => f.endsWith('.md'));
+  if (files.length === 0) {
+    return { error: 'No result files found in temp directory' };
+  }
+
+  const issueResults = [];
+
+  for (const file of files) {
+    const content = fs.readFileSync(path.join(tempDir, file), 'utf8');
+    const { meta, body } = parseFrontmatter(content);
+
+    if (meta.type === 'issue') {
+      issueResults.push({ meta, body, file });
+    }
+  }
+
+  // Sort: activity first, then by update recency
+  const withActivity = issueResults.filter(r => r.meta.has_activity === true);
+  const noActivity = issueResults.filter(r => r.meta.has_activity !== true);
+
+  // Count repos
+  const repos = new Set(issueResults.map(r => `${r.meta.owner}/${r.meta.repo}`));
+
+  // Build report
+  let report = '';
+
+  report += `## GitHub Issues Check-In — ${date}\n\n`;
+  report += `**Tracking ${issueResults.length} active issues across ${repos.size} repos**`;
+
+  // Find oldest check date for "last check"
+  const checkDates = issueResults
+    .map(r => r.meta.last_check_date)
+    .filter(Boolean)
+    .sort();
+  if (checkDates.length > 0) {
+    report += ` | Last check: ${checkDates[0]}`;
+  }
+  report += '\n\n---\n\n';
+
+  // ── Issues with Activity ──
+  if (withActivity.length > 0) {
+    report += '### Issues with Activity\n\n';
+    report += `These ${withActivity.length} issue${withActivity.length === 1 ? ' has' : 's have'} new comments or state changes since your last check.\n\n`;
+    for (const r of withActivity) {
+      report += buildIssueDetailBlock(r);
+    }
+  }
+
+  // ── No Activity ──
+  if (noActivity.length > 0) {
+    report += '### No Activity\n\n';
+    for (const r of noActivity) {
+      report += buildQuietIssueBlock(r);
+    }
+  }
+
+  // ── Upstream Status ──
+  const upstreamResults = issueResults.filter(r => {
+    const upstreamSection = extractSection(r.body, 'Upstream');
+    return upstreamSection && upstreamSection.trim() !== 'N/A' && upstreamSection.trim() !== 'None';
+  });
+  if (upstreamResults.length > 0) {
+    report += '### Upstream Status\n\n';
+    report += '| Upstream issue | State | Impact |\n';
+    report += '|----------------|-------|--------|\n';
+    for (const r of upstreamResults) {
+      const section = extractSection(r.body, 'Upstream');
+      report += `| ${r.meta.owner}/${r.meta.repo}#${r.meta.number} | ${section.trim()} |\n`;
+    }
+    report += '\n---\n\n';
+  }
+
+  // ── Summary ──
+  const actionItems = withActivity.reduce((count, r) => {
+    const steps = extractSection(r.body, 'Next Steps');
+    if (steps) count += (steps.match(/^- \[/gm) || []).length;
+    return count;
+  }, 0);
+
+  report += '### Summary\n\n';
+  report += `**Action items:** ${actionItems}\n`;
+  if (withActivity.length > 0) {
+    const needAttention = withActivity.slice(0, 3).map(
+      r => `${r.meta.owner}/${r.meta.repo}#${r.meta.number}`
+    );
+    report += `**Issues needing attention:** ${needAttention.join(', ')}\n`;
+  }
+  if (noActivity.length > 0) {
+    report += `**All quiet:** ${noActivity.length} issues with no activity\n`;
+  }
+
+  return { report, issue_count: issueResults.length, activity_count: withActivity.length };
+}
+
+function buildIssueDetailBlock(r) {
+  let block = '';
+  block += `#### ${r.meta.owner}/${r.meta.repo}#${r.meta.number} — ${r.meta.title}\n`;
+  block += `| Field | Value |\n`;
+  block += `|-------|-------|\n`;
+  block += `| **State** | ${r.meta.state || 'Open'}${r.meta.state_changed ? ' [changed]' : ''} |\n`;
+  if (r.meta.labels) {
+    const labels = Array.isArray(r.meta.labels) ? r.meta.labels.join(', ') : r.meta.labels;
+    block += `| **Labels** | ${labels} |\n`;
+  }
+  block += `| **Your role** | ${r.meta.role || 'Unknown'} |\n\n`;
+
+  // Plain English summary of where this issue stands
+  const summary = extractSection(r.body, 'Status Summary');
+  if (summary) {
+    block += `${summary.trim()}\n\n`;
+  }
+
+  // Activity section — what changed since last check
+  const activity = extractSection(r.body, 'Activity');
+  if (activity) {
+    block += `**What changed:**\n${activity}\n\n`;
+  }
+
+  // Duplicates
+  const dupes = extractSection(r.body, 'Known');
+  const newFinds = extractSection(r.body, 'New finds');
+  const hasDupes = dupes && !/^(none|no changes)\.?$/i.test(dupes.trim());
+  const hasNewFinds = newFinds && !/^(none|none found|skipped)\.?/i.test(newFinds.trim());
+  if (hasDupes || hasNewFinds) {
+    block += `**Duplicates & related:**\n`;
+    if (hasDupes) block += `- Known dupes — ${dupes.trim()}\n`;
+    if (hasNewFinds) block += `- New finds — ${newFinds.trim()}\n`;
+    block += '\n';
+  }
+
+  // What you need to do next
+  const steps = extractSection(r.body, 'Next Steps');
+  if (steps) {
+    block += `**What to do next:**\n${steps}\n\n`;
+  }
+
+  // Watch For
+  const watch = extractSection(r.body, 'Watch For');
+  if (watch) {
+    block += `**Watch for:**\n${watch}\n\n`;
+  }
+
+  block += '---\n\n';
+  return block;
+}
+
+function buildQuietIssueBlock(r) {
+  let block = '';
+  block += `#### ${r.meta.owner}/${r.meta.repo}#${r.meta.number} — ${r.meta.title}\n`;
+  const lastDate = r.meta.last_comment_date || r.meta.last_check_date || 'unknown';
+  block += `- Last activity: ${lastDate}\n`;
+
+  // Keep quiet issues compact to avoid report bloat.
+  const summary = extractSection(r.body, 'Status Summary');
+  if (summary) {
+    block += `- Status: ${firstLine(summary)}\n`;
+  }
+
+  const steps = extractSection(r.body, 'Next Steps');
+  if (steps) {
+    block += `- Next: ${firstLine(steps)}\n`;
+  }
+
+  const watch = extractSection(r.body, 'Watch For');
+  if (watch) {
+    block += `- Watch: ${firstLine(watch)}\n`;
+  }
+
+  block += '\n';
+  return block;
+}
+
+function extractSection(body, headerPattern) {
+  // Match ## Header or ### Header, capture until next ## or end
+  const re = new RegExp(
+    `^#{2,4}\\s+(?:[^\\n]*?${headerPattern}[^\\n]*)\\n([\\s\\S]*?)(?=^#{2,4}\\s|$)`,
+    'mi'
+  );
+  const match = body.match(re);
+  return match ? match[1].trim() : null;
+}
+
+function firstLine(text) {
+  if (!text) return '';
+  const line = text.split(/\r?\n/).find(l => l.trim().length > 0) || '';
+  return line.replace(/^\s*[-*]\s*/, '').trim();
+}
+
+// ─── Tracker Updater ────────────────────────────────────────────────────────
+
+function applyTrackerUpdates(trackerContent, tempDir, date) {
+  const files = fs.readdirSync(tempDir).filter(f => f.endsWith('.md'));
+  let updated = trackerContent;
+  const changes = [];
+
+  for (const file of files) {
+    const content = fs.readFileSync(path.join(tempDir, file), 'utf8');
+    const { meta, body } = parseFrontmatter(content);
+
+    if (meta.type !== 'issue') continue;
+
+    const issueKey = `${meta.owner}/${meta.repo}#${meta.number}`;
+
+    // Find this issue's section in the tracker
+    const sectionRe = new RegExp(
+      `(### ${escapeRegex(meta.owner)}/${escapeRegex(meta.repo)}#${meta.number}\\s*[—–-][^\\n]*\\n[\\s\\S]*?)(?=\\n### |\\n## |$)`,
+      'm'
+    );
+    const sectionMatch = updated.match(sectionRe);
+    if (!sectionMatch) continue;
+
+    let section = sectionMatch[1];
+    const originalSection = section;
+
+    // Update "Status as of" date and summary
+    const trackerUpdates = extractSection(body, 'Tracker Updates');
+    if (trackerUpdates) {
+      const statusLine = trackerUpdates.match(/^status_summary:\s*(.+)/m);
+      if (statusLine) {
+        section = section.replace(
+          /\*\*Status as of \d{4}-\d{2}-\d{2}:\*\*\s*.+/,
+          `**Status as of ${date}:** ${statusLine[1].trim()}`
+        );
+      }
+
+      const watchLine = trackerUpdates.match(/^what_to_check:\s*(.+)/m);
+      if (watchLine) {
+        section = section.replace(
+          /\*\*What to check\*\*:\s*.+/,
+          `**What to check:** ${watchLine[1].trim()}`
+        );
+      }
+
+      // Update or add Goal
+      const goalLine = trackerUpdates.match(/^goal:\s*(.+)/m);
+      if (goalLine) {
+        if (section.includes('**Goal:**')) {
+          section = section.replace(
+            /\*\*Goal:\*\*\s*.+/,
+            `**Goal:** ${goalLine[1].trim()}`
+          );
+        } else {
+          // Insert Goal as first field after the header line
+          section = section.replace(
+            /(### [^\n]+\n)/,
+            `$1- **Goal:** ${goalLine[1].trim()}\n`
+          );
+        }
+        changes.push(`Set Goal for ${issueKey}`);
+      }
+    }
+
+    // Handle state change: open → closed (move to Closed section)
+    if (meta.state === 'closed' && meta.state_changed) {
+      // Preserve key fields from the original section before removing it
+      const closedLines = [];
+      closedLines.push(`### ${issueKey} — ${meta.title}`);
+      closedLines.push(`- Closed as of ${date}. ${meta.close_reason || ''}`);
+
+      // Preserve Goal
+      const goalMatch = originalSection.match(/- \*\*Goal:\*\*\s*(.+)/);
+      if (goalMatch) closedLines.push(`- **Goal:** ${goalMatch[1].trim()}`);
+
+      // Preserve Role
+      const roleMatch = originalSection.match(/- \*\*Role:\*\*\s*(.+)/);
+      if (roleMatch) closedLines.push(`- **Role:** ${roleMatch[1].trim()}`);
+
+      // Preserve History section
+      const historyIdx = originalSection.indexOf('- **History:**');
+      if (historyIdx !== -1) {
+        const historyBlock = originalSection.slice(historyIdx);
+        const historyLines = historyBlock.split(/\r?\n/);
+        const preserved = [historyLines[0]]; // "- **History:**"
+        for (let hi = 1; hi < historyLines.length; hi++) {
+          if (/^\s+-\s+\*\*\d{4}-\d{2}-\d{2}:\*\*/.test(historyLines[hi])) {
+            preserved.push(historyLines[hi]);
+          } else if (/^- \*\*[A-Z]/.test(historyLines[hi]) || /^### /.test(historyLines[hi])) {
+            break;
+          }
+        }
+        // Add closing history entry
+        preserved.push(`  - **${date}:** Closed. ${meta.close_reason || ''}`);
+        closedLines.push(...preserved);
+      }
+
+      const closedEntry = closedLines.join('\n') + '\n\n';
+
+      // Remove from active
+      updated = updated.replace(originalSection, '');
+      // Add to closed
+      updated = updated.replace(
+        /(## Closed[^\n]*\n)/,
+        `$1\n${closedEntry}`
+      );
+      changes.push(`Moved ${issueKey} to Closed`);
+      continue;
+    }
+
+    // Add new duplicates
+    if (trackerUpdates) {
+      const newDupeLines = [];
+      const dupeRe = /new_duplicate:\s*#?(\d+)\s*[—–-]\s*(.+)/gm;
+      let dupeMatch;
+      while ((dupeMatch = dupeRe.exec(trackerUpdates)) !== null) {
+        newDupeLines.push(`  - **#${dupeMatch[1]}** — ${dupeMatch[2].trim()}`);
+      }
+      if (newDupeLines.length > 0) {
+        const dupeHeader = `**Duplicates found (${date}):**`;
+        if (section.includes('Duplicates found')) {
+          // Append to existing duplicates section
+          section = section.replace(
+            /(Duplicates found[^*]*\*\*:?\s*\n(?:\s+-[^\n]*\n)*)/,
+            `$1${dupeHeader}\n${newDupeLines.join('\n')}\n`
+          );
+        } else {
+          // Add before Next steps
+          section = section.replace(
+            /(\*\*Next steps)/,
+            `- ${dupeHeader}\n${newDupeLines.join('\n')}\n- $1`
+          );
+        }
+        changes.push(`Added ${newDupeLines.length} duplicates to ${issueKey}`);
+      }
+    }
+
+    // Append history entries from result file
+    if (trackerUpdates) {
+      const newHistoryEntries = [];
+      const historyRe = /^history_entry:\s*(\d{4}-\d{2}-\d{2})\s*\|\s*(.+)/gm;
+      let hm;
+      while ((hm = historyRe.exec(trackerUpdates)) !== null) {
+        newHistoryEntries.push({ date: hm[1], desc: hm[2].trim() });
+      }
+
+      if (newHistoryEntries.length > 0) {
+        const historyHeaderIdx = section.indexOf('- **History:**');
+        if (historyHeaderIdx !== -1) {
+          // History section exists — collect existing entries for dedup
+          const afterHeader = section.slice(historyHeaderIdx);
+          const existingLines = afterHeader.split(/\r?\n/);
+          const existingTexts = new Set();
+          for (const line of existingLines.slice(1)) {
+            const m = line.match(/^\s+-\s+\*\*(\d{4}-\d{2}-\d{2}):\*\*\s*(.+)/);
+            if (m) existingTexts.add(`${m[1]}|${m[2].trim()}`);
+            else if (line.match(/^- \*\*[A-Z]/) || line.match(/^### /)) break;
+          }
+
+          // Build lines to append (deduped)
+          const toAppend = newHistoryEntries
+            .filter(e => !existingTexts.has(`${e.date}|${e.desc}`))
+            .map(e => `  - **${e.date}:** ${e.desc}`);
+
+          if (toAppend.length > 0) {
+            // Find the last history bullet line position and insert after it
+            // We'll insert the new lines before the next top-level field after **History:**
+            const historyBlock = section.slice(historyHeaderIdx);
+            const historyLines = historyBlock.split(/\r?\n/);
+            let lastBulletLine = 0;
+            for (let li = 1; li < historyLines.length; li++) {
+              if (/^\s+-\s+\*\*\d{4}-\d{2}-\d{2}:\*\*/.test(historyLines[li])) {
+                lastBulletLine = li;
+              } else if (/^- \*\*[A-Z]/.test(historyLines[li]) || /^### /.test(historyLines[li])) {
+                break;
+              }
+            }
+            // Insert toAppend after lastBulletLine
+            historyLines.splice(lastBulletLine + 1, 0, ...toAppend);
+            const newHistoryBlock = historyLines.join('\n');
+            section = section.slice(0, historyHeaderIdx) + newHistoryBlock;
+            changes.push(`Appended ${toAppend.length} history entries to ${issueKey}`);
+          }
+        } else {
+          // No history section yet — insert before trailing newline / end of section
+          const insertPoint = section.lastIndexOf('\n');
+          const historyLines = ['- **History:**'];
+          for (const e of newHistoryEntries) {
+            historyLines.push(`  - **${e.date}:** ${e.desc}`);
+          }
+          section = section.slice(0, insertPoint) + '\n' + historyLines.join('\n') + section.slice(insertPoint);
+          changes.push(`Added History section with ${newHistoryEntries.length} entries to ${issueKey}`);
+        }
+      }
+    }
+
+    if (section !== originalSection) {
+      updated = updated.replace(originalSection, section);
+      changes.push(`Updated ${issueKey}`);
+    }
+  }
+
+  return { content: updated, changes };
+}
+
+function escapeRegex(str) {
+  return str.replace(/[.*+?^${}()|[\]\\]/g, '\\$&');
+}
+
+// ─── Command Implementations ────────────────────────────────────────────────
+
+function cmdCompileReport(flags) {
+  const tempDir = flags['temp-dir'];
+  const date = flags.date;
+
+  if (!tempDir || !date) {
+    console.error('Usage: compile-report --temp-dir <dir> --date <YYYY-MM-DD>');
+    process.exit(1);
+  }
+
+  if (!fs.existsSync(tempDir)) {
+    console.error(`Temp directory not found: ${tempDir}`);
+    process.exit(1);
+  }
+
+  const result = compileOverviewReport(tempDir, date);
+  if (result.error) {
+    console.error(result.error);
+    process.exit(1);
+  }
+
+  // Write report to temp dir as well for validation
+  fs.writeFileSync(path.join(tempDir, '_compiled-report.md'), result.report, 'utf8');
+
+  // Output the report to stdout
+  console.log(result.report);
+
+  // Write metadata to stderr for the agent to parse
+  console.error(JSON.stringify({
+    today: todayTag(),
+    issue_count: result.issue_count,
+    activity_count: result.activity_count,
+    report_file: path.join(tempDir, '_compiled-report.md'),
+  }));
+}
+
+function cmdUpdateTracker(flags) {
+  const trackerPath = flags.tracker;
+  const tempDir = flags['temp-dir'];
+  const date = flags.date;
+
+  if (!trackerPath || !tempDir || !date) {
+    console.error('Usage: update-tracker --tracker <path> --temp-dir <dir> --date <YYYY-MM-DD>');
+    process.exit(1);
+  }
+
+  if (!fs.existsSync(trackerPath)) {
+    console.error(`Tracker file not found: ${trackerPath}`);
+    process.exit(1);
+  }
+
+  const content = fs.readFileSync(trackerPath, 'utf8');
+  const { content: updated, changes } = applyTrackerUpdates(content, tempDir, date);
+
+  if (changes.length === 0) {
+    console.log(JSON.stringify({ today: todayTag(), updated: false, changes: [] }));
+    return;
+  }
+
+  fs.writeFileSync(trackerPath, updated, 'utf8');
+  console.log(JSON.stringify({ today: todayTag(), updated: true, changes }));
+}
+
+// ─── Async Helper ──────────────────────────────────────────────────────────
+
+function execAsync(cmd) {
+  return new Promise((resolve, reject) => {
+    exec(cmd, { maxBuffer: 10 * 1024 * 1024 }, (err, stdout, stderr) => {
+      if (err) reject(err);
+      else resolve(stdout);
+    });
+  });
+}
+
+/**
+ * Run promises in batches to avoid rate limits.
+ * @param {Array<() => Promise>} tasks - Array of functions returning promises
+ * @param {number} concurrency - Max concurrent tasks
+ * @returns {Promise<Array>} Results in same order as tasks
+ */
+async function batchRun(tasks, concurrency = 15) {
+  const results = [];
+  for (let i = 0; i < tasks.length; i += concurrency) {
+    const batch = tasks.slice(i, i + concurrency);
+    const batchResults = await Promise.all(batch.map(fn => fn()));
+    results.push(...batchResults);
+  }
+  return results;
+}
+
+// ─── Command: startup ──────────────────────────────────────────────────────
+
+async function cmdStartup(flags) {
+  const trackerPath = flags.tracker;
+  if (!trackerPath) {
+    console.error('Usage: startup --tracker <path>');
+    process.exit(1);
+  }
+
+  // Step 1: Check gh auth
+  let username = null;
+  try {
+    const authOutput = execSync('gh auth status 2>&1', { encoding: 'utf8' });
+    const userMatch = authOutput.match(/Logged in to github\.com.*account\s+(\S+)/i)
+      || authOutput.match(/Logged in to github\.com\s+as\s+(\S+)/i)
+      || authOutput.match(/account\s+(\S+)/i);
+    if (userMatch) username = userMatch[1];
+  } catch (err) {
+    console.log(JSON.stringify({
+      script_ok: true, auth: false, today: todayTag(),
+      error: 'Not authenticated with GitHub. Run `gh auth login` in your terminal to fix this.'
+    }));
+    return;
+  }
+
+  if (!username) {
+    // Try alternate extraction from gh api
+    try {
+      username = execSync('gh api user --jq .login', { encoding: 'utf8' }).trim();
+    } catch (_) {
+      console.log(JSON.stringify({
+        script_ok: true, auth: false, today: todayTag(),
+        error: 'Not authenticated with GitHub. Run `gh auth login` in your terminal to fix this.'
+      }));
+      return;
+    }
+  }
+
+  // Step 2: Parse tracker (or set empty defaults if no tracker)
+  const thirtyDaysAgo = new Date(Date.now() - 30 * 24 * 60 * 60 * 1000)
+    .toISOString().split('T')[0];
+
+  let trackerExists = false;
+  let parsed = { username: null, active_issues: [], closed_issues: [], raw: '' };
+
+  if (fs.existsSync(trackerPath)) {
+    const content = fs.readFileSync(trackerPath, 'utf8');
+    if (content.trim()) {
+      trackerExists = true;
+      parsed = parseTrackerFile(content);
+      // Default null last_check_date to 30 days ago
+      for (const issue of parsed.active_issues) {
+        if (!issue.last_check_date) issue.last_check_date = thirtyDaysAgo;
+      }
+    }
+  }
+
+  const oldestCheckDate = parsed.active_issues
+    .map(i => i.last_check_date)
+    .filter(Boolean)
+    .sort()[0] || thirtyDaysAgo;
+
+  const allTrackedNumbers = [
+    ...parsed.active_issues.map(i => `${i.owner}/${i.repo}#${i.number}`),
+    ...parsed.closed_issues.map(i => `${i.owner}/${i.repo}#${i.number}`),
+  ];
+
+  // Step 3: Run two gh api search queries in parallel (always, even without tracker)
+  const searchDate = oldestCheckDate;
+  const openQuery = `search/issues?q=involves:${username}+is:open+updated:>${searchDate}&per_page=100`;
+  const closedQuery = `search/issues?q=involves:${username}+is:closed+closed:>${thirtyDaysAgo}&per_page=50`;
+
+  let openResults = [];
+  let closedResults = [];
+  let searchErrors = [];
+
+  // Run searches individually so one failure doesn't block the other
+  try {
+    const openRaw = await execAsync(`gh api "${openQuery}" --jq ".items"`);
+    openResults = JSON.parse(openRaw || '[]');
+  } catch (err) {
+    searchErrors.push(`open search failed: ${err.message}`);
+    openResults = [];
+  }
+
+  try {
+    const closedRaw = await execAsync(`gh api "${closedQuery}" --jq ".items"`);
+    closedResults = JSON.parse(closedRaw || '[]');
+  } catch (err) {
+    searchErrors.push(`closed search failed: ${err.message}`);
+    closedResults = [];
+  }
+
+  // Step 4: Identify new, reopened, recently closed
+  function issueKey(item) {
+    const urlParts = (item.repository_url || '').split('/');
+    const owner = urlParts[urlParts.length - 2];
+    const repo = urlParts[urlParts.length - 1];
+    return `${owner}/${repo}#${item.number}`;
+  }
+
+  function issueInfo(item) {
+    const urlParts = (item.repository_url || '').split('/');
+    return {
+      owner: urlParts[urlParts.length - 2],
+      repo: urlParts[urlParts.length - 1],
+      number: item.number,
+      title: item.title,
+      state: item.state,
+      updated_at: item.updated_at,
+      created_at: item.created_at,
+      html_url: item.html_url,
+      labels: (item.labels || []).map(l => l.name),
+      user: item.user ? item.user.login : null,
+    };
+  }
+
+  const trackedSet = new Set(allTrackedNumbers);
+  const closedNumbers = new Set(
+    parsed.closed_issues.map(i => `${i.owner}/${i.repo}#${i.number}`)
+  );
+
+  const newIssues = openResults
+    .filter(item => !trackedSet.has(issueKey(item)))
+    .map(issueInfo);
+
+  const reopenedIssues = openResults
+    .filter(item => closedNumbers.has(issueKey(item)))
+    .map(issueInfo);
+
+  const recentlyClosed = closedResults.map(issueInfo);
+
+  // Create temp directory for this session
+  const prefix = 'giu-checkin-';
+  const tempDir = fs.mkdtempSync(path.join(os.tmpdir(), prefix));
+
+  console.log(JSON.stringify({
+    script_ok: true,
+    auth: true,
+    today: todayTag(),
+    username,
+    temp_dir: tempDir,
+    tracker_exists: trackerExists,
+    tracker_path: trackerPath,
+    config: parsed.config,
+    tracker_data: {
+      username: parsed.username,
+      active_issues: parsed.active_issues,
+      closed_issues: parsed.closed_issues,
+      all_tracked_numbers: allTrackedNumbers,
+      oldest_check_date: oldestCheckDate,
+      raw: parsed.raw,
+    },
+    new_issues: newIssues,
+    reopened_issues: reopenedIssues,
+    recently_closed: recentlyClosed,
+    search_errors: searchErrors,
+  }, null, 2));
+}
+
+// ─── Command: fetch-issues ─────────────────────────────────────────────────
+
+async function cmdFetchIssues(flags) {
+  const tempDir = flags['temp-dir'];
+  const issuesFile = flags.issues;
+
+  if (!tempDir || !issuesFile) {
+    console.error('Usage: fetch-issues --temp-dir <dir> --issues <json-file>');
+    process.exit(1);
+  }
+
+  if (!fs.existsSync(issuesFile)) {
+    console.error(`Issues file not found: ${issuesFile}`);
+    process.exit(1);
+  }
+
+  const issues = JSON.parse(fs.readFileSync(issuesFile, 'utf8'));
+  const errors = [];
+  const files = [];
+
+  // Build all fetch tasks
+  const allTasks = [];
+  const taskMap = []; // Maps task index to { issueIdx, type }
+
+  // Fetch raw JSON from gh api (no --jq — process in Node to avoid Windows quoting issues)
+  function ghFetch(endpoint) {
+    return execAsync(`gh api ${endpoint}`)
+      .then(raw => JSON.parse(raw || '{}'))
+      .catch(e => ({ _error: e.message }));
+  }
+
+  for (let i = 0; i < issues.length; i++) {
+    const iss = issues[i];
+    const { owner, repo, number, last_check_date, known_dupes, upstream } = iss;
+    const issueEndpoint = `repos/${owner}/${repo}/issues/${number}`;
+
+    // Issue data (single call — extract metadata + body + author in Node)
+    allTasks.push(() => ghFetch(issueEndpoint));
+    taskMap.push({ issueIdx: i, type: 'issue' });
+
+    // Comments (all of them — filter by date in Node)
+    allTasks.push(() => execAsync(`gh api ${issueEndpoint}/comments --paginate`)
+      .then(raw => JSON.parse(raw || '[]'))
+      .catch(e => ({ _error: e.message })));
+    taskMap.push({ issueIdx: i, type: 'comments' });
+
+    // Known dupe state checks
+    const dupes = known_dupes || [];
+    for (const dupe of dupes) {
+      const dupeMatch = String(dupe).match(/(?:([^/]+)\/([^#]+))?#?(\d+)/);
+      if (dupeMatch) {
+        const dOwner = dupeMatch[1] || owner;
+        const dRepo = dupeMatch[2] || repo;
+        const dNumber = dupeMatch[3];
+        allTasks.push(() => ghFetch(`repos/${dOwner}/${dRepo}/issues/${dNumber}`));
+        taskMap.push({ issueIdx: i, type: 'dupe', dupeKey: `${dOwner}/${dRepo}#${dNumber}` });
+      }
+    }
+
+    // Upstream check
+    if (upstream) {
+      const uMatch = String(upstream).match(/([^/]+)\/([^#]+)#(\d+)/);
+      if (uMatch) {
+        allTasks.push(() => ghFetch(`repos/${uMatch[1]}/${uMatch[2]}/issues/${uMatch[3]}`));
+        taskMap.push({ issueIdx: i, type: 'upstream' });
+      }
+    }
+  }
+
+  // Run all tasks with batching (max 15 concurrent)
+  const results = await batchRun(allTasks, 15);
+
+  // Organize results by issue — process raw API responses in Node
+  const issueData = issues.map(() => ({
+    metadata: null,
+    body: null,
+    comments: null,
+    dupe_states: {},
+    upstream_state: null,
+    cross_references: [],
+    urls: [],
+  }));
+
+  for (let t = 0; t < results.length; t++) {
+    const { issueIdx, type, dupeKey } = taskMap[t];
+    const raw = results[t];
+
+    switch (type) {
+      case 'issue': {
+        // Extract metadata and body from the single issue response
+        if (raw && !raw._error) {
+          issueData[issueIdx].metadata = {
+            state: raw.state,
+            labels: (raw.labels || []).map(l => l.name),
+            comments: raw.comments,
+            updated: raw.updated_at,
+            created: raw.created_at,
+            html_url: raw.html_url,
+          };
+          issueData[issueIdx].body = {
+            title: raw.title,
+            body: raw.body,
+            author: raw.user ? raw.user.login : null,
+          };
+        } else {
+          issueData[issueIdx].metadata = raw;
+          issueData[issueIdx].body = raw;
+        }
+        break;
+      }
+      case 'comments': {
+        // Filter comments by last_check_date in Node
+        const iss = issues[issueIdx];
+        if (Array.isArray(raw)) {
+          const filtered = iss.last_check_date
+            ? raw.filter(c => c.created_at > iss.last_check_date)
+            : raw;
+          issueData[issueIdx].comments = filtered.map(c => ({
+            author: c.user ? c.user.login : null,
+            date: c.created_at ? c.created_at.split('T')[0] : null,
+            body: c.body,
+          }));
+        } else {
+          issueData[issueIdx].comments = raw;
+        }
+        break;
+      }
+      case 'dupe': {
+        if (raw && !raw._error) {
+          issueData[issueIdx].dupe_states[dupeKey] = {
+            state: raw.state,
+            updated: raw.updated_at,
+          };
+        } else {
+          issueData[issueIdx].dupe_states[dupeKey] = raw;
+        }
+        break;
+      }
+      case 'upstream': {
+        if (raw && !raw._error) {
+          issueData[issueIdx].upstream_state = {
+            state: raw.state,
+            labels: (raw.labels || []).map(l => l.name),
+            updated: raw.updated_at,
+          };
+        } else {
+          issueData[issueIdx].upstream_state = raw;
+        }
+        break;
+      }
+    }
+  }
+
+  // Post-process: extract cross_references and urls from body + comments
+  for (let i = 0; i < issues.length; i++) {
+    const data = issueData[i];
+    const allText = collectText(data);
+
+    // Extract #NUMBER and owner/repo#NUMBER patterns
+    const crossRefs = new Set();
+    const refRe = /(?:([a-zA-Z0-9_.-]+\/[a-zA-Z0-9_.-]+))?#(\d+)/g;
+    let m;
+    while ((m = refRe.exec(allText)) !== null) {
+      crossRefs.add(m[1] ? `${m[1]}#${m[2]}` : `#${m[2]}`);
+    }
+    data.cross_references = [...crossRefs];
+
+    // Extract URLs
+    const urlSet = new Set();
+    const urlRe = /https?:\/\/[^\s)>\]"']+/g;
+    while ((m = urlRe.exec(allText)) !== null) {
+      urlSet.add(m[0].replace(/[.,;:]+$/, ''));
+    }
+    data.urls = [...urlSet];
+
+    // Write raw JSON per issue
+    const iss = issues[i];
+    const outFile = path.join(tempDir, `raw-${iss.owner}-${iss.repo}-${iss.number}.json`);
+    try {
+      fs.writeFileSync(outFile, JSON.stringify(data, null, 2), 'utf8');
+      files.push(outFile);
+    } catch (err) {
+      errors.push(`Failed to write ${outFile}: ${err.message}`);
+    }
+  }
+
+  console.log(JSON.stringify({ today: todayTag(), fetched: files.length, files, errors }));
+}
+
+/**
+ * Collect all text content from issue data for cross-reference/URL extraction.
+ */
+function collectText(data) {
+  const parts = [];
+  if (data.body) {
+    if (typeof data.body === 'string') parts.push(data.body);
+    else if (data.body.body) parts.push(data.body.body);
+  }
+  if (data.comments) {
+    if (typeof data.comments === 'string') parts.push(data.comments);
+    else if (Array.isArray(data.comments)) {
+      for (const c of data.comments) {
+        if (typeof c === 'string') parts.push(c);
+        else if (c && c.body) parts.push(c.body);
+      }
+    }
+  }
+  return parts.join('\n');
+}
+
+// ─── Command: build-tracker ────────────────────────────────────────────────
+
+function cmdBuildTracker(flags) {
+  const tempDir = flags['temp-dir'];
+  const templatePath = flags.template;
+  const username = flags.username;
+  const trackerPath = flags.tracker;
+  const closedJsonPath = flags['closed-json'];
+  const date = flags.date || localDate();
+
+  if (!tempDir || !templatePath || !username || !trackerPath) {
+    console.error('Usage: build-tracker --temp-dir <dir> --template <path> --username <name> --tracker <path> [--closed-json <path>]');
+    process.exit(1);
+  }
+
+  // Step 1: Read template
+  if (!fs.existsSync(templatePath)) {
+    console.error(`Template not found: ${templatePath}`);
+    process.exit(1);
+  }
+  let tracker = fs.readFileSync(templatePath, 'utf8');
+
+  // Step 2: Replace USERNAME_HERE
+  tracker = tracker.replace(/USERNAME_HERE/g, username);
+
+  // Step 3: Read all issue-*.md result files from temp dir
+  if (!fs.existsSync(tempDir)) {
+    console.error(`Temp directory not found: ${tempDir}`);
+    process.exit(1);
+  }
+
+  const resultFiles = fs.readdirSync(tempDir).filter(f =>
+    f.startsWith('issue-') && f.endsWith('.md')
+  );
+
+  const entries = [];
+  for (const file of resultFiles) {
+    const content = fs.readFileSync(path.join(tempDir, file), 'utf8');
+    const { meta, body } = parseFrontmatter(content);
+    if (meta.type !== 'issue') continue;
+
+    // Build tracker entry from result file
+    const entry = buildTrackerEntry(meta, body, date);
+    entries.push(entry);
+  }
+
+  // Step 4: Insert active entries into tracker
+  const activeEntriesText = entries.join('\n');
+  tracker = tracker.replace(
+    /(## Active Issues[^\n]*\n)([\s\S]*?)(## Closed)/,
+    `$1\n${activeEntriesText}\n$3`
+  );
+
+  // Step 5: Handle closed issues if provided
+  let closedCount = 0;
+  if (closedJsonPath && fs.existsSync(closedJsonPath)) {
+    const closedIssues = JSON.parse(fs.readFileSync(closedJsonPath, 'utf8'));
+    closedCount = closedIssues.length;
+    let closedText = '';
+    for (const ci of closedIssues) {
+      closedText += `### ${ci.owner}/${ci.repo}#${ci.number} — ${ci.title}\n`;
+      closedText += `- ${ci.resolution || 'Closed.'}\n\n`;
+    }
+    if (closedText) {
+      tracker = tracker.replace(
+        /(## Closed[^\n]*\n)/,
+        `$1\n${closedText}`
+      );
+    }
+  }
+
+  // Step 6: Write tracker
+  fs.writeFileSync(trackerPath, tracker, 'utf8');
+
+  console.log(JSON.stringify({
+    today: todayTag(),
+    written: true,
+    path: trackerPath,
+    active_count: entries.length,
+    closed_count: closedCount,
+  }));
+}
+
+/**
+ * Build a tracker entry string from result file frontmatter and body.
+ */
+function buildTrackerEntry(meta, body, date) {
+  const lines = [];
+  lines.push(`### ${meta.owner}/${meta.repo}#${meta.number} — ${meta.title}`);
+
+  // Goal — from tracker updates or frontmatter
+  const trackerUpdatesForGoal = extractSection(body, 'Tracker Updates') || '';
+  const goalLine = trackerUpdatesForGoal.match(/^goal:\s*(.+)/m);
+  if (goalLine) {
+    lines.push(`- **Goal:** ${goalLine[1].trim()}`);
+  }
+
+  // Role
+  lines.push(`- **Role:** ${meta.role || 'Unknown'}`);
+
+  // Filed
+  if (meta.filed) {
+    lines.push(`- **Filed:** ${meta.filed}`);
+  }
+
+  // Status — from ## Status Summary + labels
+  const statusSummary = extractSection(body, 'Status Summary') || 'Open.';
+  const labels = meta.labels
+    ? (Array.isArray(meta.labels) ? meta.labels.join(', ') : meta.labels)
+    : '';
+  const trackerUpdates = extractSection(body, 'Tracker Updates') || '';
+  const statusLine = trackerUpdates.match(/^status_summary:\s*(.+)/m);
+  const statusText = statusLine
+    ? statusLine[1].trim()
+    : `${meta.state || 'Open'}. Labels: ${labels}. ${statusSummary.split('\n')[0]}`;
+  const dateStr = date || localDate();
+  lines.push(`- **Status as of ${dateStr}:** ${statusText}`);
+
+  // What to check — from ## Watch For or tracker updates
+  const watchLine = trackerUpdates.match(/^what_to_check:\s*(.+)/m);
+  const watchFor = watchLine
+    ? watchLine[1].trim()
+    : (extractSection(body, 'Watch For') || 'Monitor for updates.').split('\n')[0].replace(/^-\s*/, '');
+  lines.push(`- **What to check:** ${watchFor}`);
+
+  // Related — from ## Cross-References
+  const crossRefs = extractSection(body, 'Cross-References');
+  if (crossRefs && crossRefs.trim() !== 'None' && crossRefs.trim() !== 'None found.') {
+    lines.push(`- **Related:** ${crossRefs.split('\n')[0].trim()}`);
+  }
+
+  // Upstream
+  const upstream = extractSection(body, 'Upstream');
+  if (upstream && upstream.trim() !== 'N/A' && upstream.trim() !== 'None') {
+    lines.push(`- **Upstream:** ${upstream.split('\n')[0].trim()}`);
+  }
+
+  // Key Context -> workaround / future
+  const keyContext = extractSection(body, 'Key Context');
+  if (keyContext && keyContext.trim() !== 'N/A') {
+    // Check for workaround
+    const workaroundMatch = keyContext.match(/[Ww]orkaround:?\s*(.+)/);
+    if (workaroundMatch) {
+      lines.push(`- **Workaround:** ${workaroundMatch[1].trim()}`);
+    }
+  }
+
+  // Duplicates from ## Duplicates and Related
+  const dupeSection = extractSection(body, 'Duplicates and Related')
+    || extractSection(body, 'Known');
+  if (dupeSection) {
+    const dupeLines = dupeSection.split('\n').filter(l => l.match(/^\s*-?\s*[#*]/));
+    if (dupeLines.length > 0) {
+      lines.push(`- **Duplicates found (${dateStr}):**`);
+      for (const dl of dupeLines) {
+        const cleaned = dl.replace(/^[\s-]*/, '  - ');
+        lines.push(cleaned);
+      }
+    }
+  }
+
+  // Next steps
+  const nextSteps = extractSection(body, 'Next Steps');
+  if (nextSteps && nextSteps.trim() !== 'None' && nextSteps.trim() !== 'None — no action needed.') {
+    lines.push(`- **Next steps (now):** ${nextSteps.split('\n')[0].replace(/^-\s*\[.\]\s*/, '').trim()}`);
+  } else {
+    lines.push(`- **Next steps (now):** None — no action needed.`);
+  }
+
+  // Future — from Key Context
+  if (keyContext && keyContext.trim() !== 'N/A') {
+    lines.push(`- **Future:** ${keyContext.split('\n')[0].replace(/^-\s*/, '').trim()}`);
+  }
+
+  // History — from history_entry lines in Tracker Updates
+  const historyEntries = [];
+  if (trackerUpdates) {
+    const historyRe = /^history_entry:\s*(\d{4}-\d{2}-\d{2})\s*\|\s*(.+)/gm;
+    let hm;
+    while ((hm = historyRe.exec(trackerUpdates)) !== null) {
+      historyEntries.push(`  - **${hm[1]}:** ${hm[2].trim()}`);
+    }
+  }
+  if (historyEntries.length === 0) {
+    // Default entry on initial build
+    historyEntries.push(`  - **${dateStr}:** Added to tracker`);
+  }
+  lines.push(`- **History:**`);
+  for (const he of historyEntries) {
+    lines.push(he);
+  }
+
+  lines.push('');
+  return lines.join('\n');
+}