job-forge 2.0.0

package/merge-tracker.mjs

@@ -0,0 +1,377 @@
+#!/usr/bin/env node
+/**
+ * merge-tracker.mjs — Merge batch tracker additions into the application tracker
+ *
+ * Supports both layouts:
+ *   - Day-based: data/applications/YYYY-MM-DD.md (preferred)
+ *   - Single-file: data/applications.md or applications.md (legacy)
+ *
+ * Handles multiple TSV formats:
+ * - 9-col: num\tdate\tcompany\trole\tstatus\tscore\tpdf\treport\tnotes
+ * - 8-col: num\tdate\tcompany\trole\tstatus\tscore\tpdf\treport (no notes)
+ * - Pipe-delimited (markdown table row): | col | col | ... |
+ *
+ * Dedup: company normalized + role fuzzy match + report number match
+ * If duplicate with higher score → update in-place, update report link
+ * Validates status against templates/states.yml when present (else built-in labels)
+ *
+ * Run: node merge-tracker.mjs [--dry-run] [--verify]   (from repo root)
+ */
+
+import { readFileSync, writeFileSync, readdirSync, mkdirSync, renameSync, existsSync } from 'fs';
+import { join, relative, dirname } from 'path';
+import { fileURLToPath } from 'url';
+import {
+  PROJECT_DIR, DATA_APPS_DIR, DATA_APPS_FILE, ROOT_APPS_FILE,
+  usesDayFiles, ensureDayDir, getHeader, formatAppLine, parseAppLine,
+  readAllEntries, writeToDayFiles, listDayFiles, dayFilePath,
+} from './tracker-lib.mjs';
+
+const ADDITIONS_DIR = join(PROJECT_DIR, 'batch/tracker-additions');
+const MERGED_DIR = join(ADDITIONS_DIR, 'merged');
+const DRY_RUN = process.argv.includes('--dry-run');
+const VERIFY = process.argv.includes('--verify');
+
+if (process.argv.includes('--help') || process.argv.includes('-h')) {
+  console.log(`merge-tracker.mjs — merge batch/tracker-additions/*.tsv into the tracker
+
+Supports day-based (data/applications/YYYY-MM-DD.md) and single-file layouts.
+Moves processed files to batch/tracker-additions/merged/.
+
+Usage:
+  node merge-tracker.mjs [--dry-run] [--verify]
+  npm run merge [-- --dry-run] [--verify]
+
+Options:
+  --dry-run    Show actions without writing the tracker or moving TSVs
+  --verify     After merge, run verify-pipeline.mjs (ignored with --dry-run)
+
+If the tracker file is missing but TSVs exist, creates the tracker
+with an empty table header. If batch/tracker-additions/ is missing or has
+no .tsv files, exits successfully with nothing to do.
+
+Run from the repository root.`);
+  process.exit(0);
+}
+
+const STATES_FILE = existsSync(join(PROJECT_DIR, 'templates/states.yml'))
+  ? join(PROJECT_DIR, 'templates/states.yml')
+  : join(PROJECT_DIR, 'states.yml');
+
+function loadCanonicalLabelsFromStatesYaml(filePath) {
+  if (!existsSync(filePath)) return null;
+  const text = readFileSync(filePath, 'utf-8');
+  const labels = [];
+  for (const line of text.split('\n')) {
+    const m = line.match(/^\s+label:\s*(.+)$/);
+    if (!m) continue;
+    let v = m[1].trim().replace(/^['"]|['"]$/g, '');
+    if (v) labels.push(v);
+  }
+  return labels.length > 0 ? labels : null;
+}
+
+const CANONICAL_STATES = loadCanonicalLabelsFromStatesYaml(STATES_FILE) || [
+  'Evaluated', 'Applied', 'Contacted', 'Responded', 'Interview', 'Offer', 'Rejected', 'Discarded', 'SKIP',
+];
+
+function validateStatus(status) {
+  const clean = status.replace(/\*\*/g, '').replace(/\s+\d{4}-\d{2}-\d{2}.*$/, '').trim();
+  const lower = clean.toLowerCase();
+
+  for (const valid of CANONICAL_STATES) {
+    if (valid.toLowerCase() === lower) return valid;
+  }
+
+  const aliases = {
+    'hold': 'Evaluated',
+    'applied': 'Applied', 'sent': 'Applied',
+    'skip': 'SKIP',
+  };
+
+  if (aliases[lower]) return aliases[lower];
+
+  if (/^(dup(licate)?|repost)/i.test(lower)) return 'Discarded';
+
+  console.warn(`⚠️  Non-canonical status "${status}" → defaulting to "Evaluated"`);
+  return 'Evaluated';
+}
+
+function normalizeCompany(name) {
+  return name.toLowerCase().replace(/[^a-z0-9]/g, '');
+}
+
+function roleFuzzyMatch(a, b) {
+  const wordsA = a.toLowerCase().split(/\s+/).filter(w => w.length > 3);
+  const wordsB = b.toLowerCase().split(/\s+/).filter(w => w.length > 3);
+  const overlap = wordsA.filter(w => wordsB.some(wb => wb.includes(w) || w.includes(wb)));
+  return overlap.length >= 2;
+}
+
+function extractReportNum(reportStr) {
+  const m = reportStr.match(/\[(\d+)\]/);
+  return m ? parseInt(m[1]) : null;
+}
+
+function parseScore(s) {
+  const m = s.replace(/\*\*/g, '').match(/([\d.]+)/);
+  return m ? parseFloat(m[1]) : 0;
+}
+
+/**
+ * Parse a TSV file content into a structured addition object.
+ */
+function parseTsvContent(content, filename) {
+  content = content.trim();
+  if (!content) return null;
+
+  let parts;
+  let addition;
+
+  if (content.startsWith('|')) {
+    parts = content.split('|').map(s => s.trim()).filter(Boolean);
+    if (parts.length < 8) {
+      console.warn(`⚠️  Skipping malformed pipe-delimited ${filename}: ${parts.length} fields`);
+      return null;
+    }
+    addition = {
+      num: parseInt(parts[0]),
+      date: parts[1],
+      company: parts[2],
+      role: parts[3],
+      score: parts[4],
+      status: validateStatus(parts[5]),
+      pdf: parts[6],
+      report: parts[7],
+      notes: parts[8] || '',
+    };
+  } else {
+    parts = content.split('\t');
+    if (parts.length < 8) {
+      console.warn(`⚠️  Skipping malformed TSV ${filename}: ${parts.length} fields`);
+      return null;
+    }
+
+    const col4 = parts[4].trim();
+    const col5 = parts[5].trim();
+    const col4LooksLikeScore = /^\d+\.?\d*\/5$/.test(col4) || col4 === 'N/A' || col4 === 'DUP';
+    const col5LooksLikeScore = /^\d+\.?\d*\/5$/.test(col5) || col5 === 'N/A' || col5 === 'DUP';
+    const col4LooksLikeStatus = /^(evaluated|applied|contacted|responded|interview|offer|rejected|discarded|skip|duplicate|repost|hold)/i.test(col4);
+    const col5LooksLikeStatus = /^(evaluated|applied|contacted|responded|interview|offer|rejected|discarded|skip|duplicate|repost|hold)/i.test(col5);
+
+    let statusCol, scoreCol;
+    if (col4LooksLikeStatus && !col4LooksLikeScore) {
+      statusCol = col4; scoreCol = col5;
+    } else if (col4LooksLikeScore && col5LooksLikeStatus) {
+      statusCol = col5; scoreCol = col4;
+    } else if (col5LooksLikeScore && !col4LooksLikeScore) {
+      statusCol = col4; scoreCol = col5;
+    } else {
+      statusCol = col4; scoreCol = col5;
+    }
+
+    addition = {
+      num: parseInt(parts[0]),
+      date: parts[1],
+      company: parts[2],
+      role: parts[3],
+      status: validateStatus(statusCol),
+      score: scoreCol,
+      pdf: parts[6],
+      report: parts[7],
+      notes: parts[8] || '',
+    };
+  }
+
+  if (isNaN(addition.num) || addition.num === 0) {
+    console.warn(`⚠️  Skipping ${filename}: invalid entry number`);
+    return null;
+  }
+
+  return addition;
+}
+
+// ---- Main ----
+
+if (!existsSync(ADDITIONS_DIR)) {
+  console.log('✅ No pending additions to merge.');
+  process.exit(0);
+}
+
+const tsvFiles = readdirSync(ADDITIONS_DIR).filter(f => f.endsWith('.tsv'));
+if (tsvFiles.length === 0) {
+  console.log('✅ No pending additions to merge.');
+  process.exit(0);
+}
+
+// Initialize tracker
+const layout = usesDayFiles() ? 'day' : 'single';
+let appLines;
+let existingApps;
+let maxNum;
+
+if (layout === 'day') {
+  ensureDayDir();
+  ({ entries: existingApps, maxNum } = readAllEntries());
+} else {
+  // Single-file mode
+  const APPS_FILE = existsSync(DATA_APPS_FILE) ? DATA_APPS_FILE : ROOT_APPS_FILE;
+  const appsDisplay = relative(PROJECT_DIR, APPS_FILE).replace(/\\/g, '/');
+
+  if (!existsSync(APPS_FILE)) {
+    if (DRY_RUN) {
+      console.log('(dry-run) Would create data/applications.md with empty tracker header.');
+    } else {
+      console.log('No tracker file yet; creating data/applications.md with empty header.');
+      mkdirSync(join(PROJECT_DIR, 'data'), { recursive: true });
+      writeFileSync(DATA_APPS_FILE, getHeader() + '\n', 'utf-8');
+    }
+  }
+
+  const filePath = existsSync(DATA_APPS_FILE) ? DATA_APPS_FILE : ROOT_APPS_FILE;
+  appLines = readFileSync(filePath, 'utf-8').split('\n');
+  existingApps = [];
+  maxNum = 0;
+  for (const line of appLines) {
+    if (line.startsWith('|') && !line.includes('---') && !line.includes('Company')) {
+      const app = parseAppLine(line);
+      if (app) {
+        existingApps.push(app);
+        if (app.num > maxNum) maxNum = app.num;
+      }
+    }
+  }
+}
+
+const appsDisplay = layout === 'day' ? relative(PROJECT_DIR, DATA_APPS_DIR) : relative(PROJECT_DIR, existsSync(DATA_APPS_FILE) ? DATA_APPS_FILE : ROOT_APPS_FILE);
+console.log(`📊 ${appsDisplay}: ${existingApps.length} existing entries, max #${maxNum}`);
+
+tsvFiles.sort((a, b) => {
+  const numA = parseInt(a.replace(/\D/g, '')) || 0;
+  const numB = parseInt(b.replace(/\D/g, '')) || 0;
+  return numA - numB;
+});
+
+console.log(`📥 Found ${tsvFiles.length} pending additions`);
+
+let added = 0;
+let updated = 0;
+let skipped = 0;
+const newEntries = [];
+
+for (const file of tsvFiles) {
+  const content = readFileSync(join(ADDITIONS_DIR, file), 'utf-8').trim();
+  const addition = parseTsvContent(content, file);
+  if (!addition) { skipped++; continue; }
+
+  const reportNum = extractReportNum(addition.report);
+  let duplicate = null;
+
+  if (reportNum) {
+    duplicate = existingApps.find(app => {
+      const existingReportNum = extractReportNum(app.report);
+      return existingReportNum === reportNum;
+    });
+  }
+
+  if (!duplicate) {
+    duplicate = existingApps.find(app => app.num === addition.num);
+  }
+
+  if (!duplicate) {
+    const normCompany = normalizeCompany(addition.company);
+    duplicate = existingApps.find(app => {
+      if (normalizeCompany(app.company) !== normCompany) return false;
+      return roleFuzzyMatch(addition.role, app.role);
+    });
+  }
+
+  if (duplicate) {
+    const newScore = parseScore(addition.score);
+    const oldScore = parseScore(duplicate.score);
+
+    if (newScore > oldScore) {
+      console.log(`🔄 Update: #${duplicate.num} ${addition.company} — ${addition.role} (${oldScore}→${newScore})`);
+
+      if (layout === 'day') {
+        // Update in existing entries list for later write
+        duplicate.date = addition.date;
+        duplicate.company = addition.company;
+        duplicate.role = addition.role;
+        duplicate.score = addition.score;
+        duplicate.report = addition.report;
+        duplicate.notes = `Re-eval ${addition.date} (${oldScore}→${newScore}). ${addition.notes}`;
+      } else {
+        const lineIdx = appLines.indexOf(duplicate.raw);
+        if (lineIdx >= 0) {
+          appLines[lineIdx] = `| ${duplicate.num} | ${addition.date} | ${addition.company} | ${addition.role} | ${addition.score} | ${duplicate.status} | ${duplicate.pdf} | ${addition.report} | Re-eval ${addition.date} (${oldScore}→${newScore}). ${addition.notes} |`;
+        }
+      }
+      updated++;
+    } else {
+      console.log(`⏭️  Skip: ${addition.company} — ${addition.role} (existing #${duplicate.num} ${oldScore} >= new ${newScore})`);
+      skipped++;
+    }
+  } else {
+    const entryNum = addition.num > maxNum ? addition.num : ++maxNum;
+    if (addition.num > maxNum) maxNum = addition.num;
+
+    newEntries.push({
+      ...addition,
+      num: entryNum,
+    });
+    added++;
+    console.log(`➕ Add #${entryNum}: ${addition.company} — ${addition.role} (${addition.score})`);
+  }
+}
+
+// Write new entries
+if (!DRY_RUN) {
+  if (layout === 'day') {
+    // Merge new entries into existing, then write day files
+    existingApps.push(...newEntries);
+    writeToDayFiles(existingApps);
+  } else {
+    // Single-file: insert new lines after header
+    if (newEntries.length > 0) {
+      let insertIdx = -1;
+      for (let i = 0; i < appLines.length; i++) {
+        if (appLines[i].includes('---') && appLines[i].startsWith('|')) {
+          insertIdx = i + 1;
+          break;
+        }
+      }
+      if (insertIdx >= 0) {
+        appLines.splice(insertIdx, 0, ...newEntries.map(formatAppLine));
+      }
+    }
+
+    const APPS_FILE = existsSync(DATA_APPS_FILE) ? DATA_APPS_FILE : ROOT_APPS_FILE;
+    writeFileSync(APPS_FILE, appLines.join('\n'));
+  }
+
+  // Move processed files to merged/
+  if (!existsSync(MERGED_DIR)) mkdirSync(MERGED_DIR, { recursive: true });
+  for (const file of tsvFiles) {
+    renameSync(join(ADDITIONS_DIR, file), join(MERGED_DIR, file));
+  }
+  console.log(`\n✅ Moved ${tsvFiles.length} TSVs to merged/`);
+}
+
+console.log(`\n📊 Summary: +${added} added, 🔄${updated} updated, ⏭️${skipped} skipped`);
+if (DRY_RUN) console.log('(dry-run — no changes written)');
+
+// Optional verify — resolve verify-pipeline.mjs relative to this file (works whether
+// installed as a package in node_modules or run from the repo root).
+if (VERIFY && !DRY_RUN) {
+  console.log('\n--- Running verification ---');
+  const { execSync } = await import('child_process');
+  const { fileURLToPath } = await import('url');
+  const { dirname } = await import('path');
+  const here = dirname(fileURLToPath(import.meta.url));
+  try {
+    execSync(`node ${join(here, 'verify-pipeline.mjs')}`, { stdio: 'inherit' });
+  } catch (e) {
+    process.exit(1);
+  }
+}

package/modes/README.md

@@ -0,0 +1,30 @@
+# Modes
+
+Markdown prompts used with opencode together with the root [`OPENCODE.md`](../OPENCODE.md). Each file aligns with a `/job-forge …` entry point or shared behavior described there.
+
+- **`_shared.md`** — Archetypes, scoring dimensions, negotiation scaffolding. Edit this first when you change how offers are classified or weighted.
+- **Per-command files** — Each `*.md` here pairs with a `/job-forge …` entry in [`OPENCODE.md`](../OPENCODE.md). How modes connect to batch, tracker, and scripts is spelled out in [**Architecture — Modes**](../docs/ARCHITECTURE.md#modes-modes).
+
+| File | Role |
+|------|------|
+| [`_shared.md`](_shared.md) | Shared archetypes, scoring, negotiation scaffolding |
+| [`auto-pipeline.md`](auto-pipeline.md) | Default path when the user pastes a JD or URL — full evaluate → report → PDF → tracker flow |
+| [`offer.md`](offer.md) | Explicit full evaluation (blocks A–F) for a single offer |
+| [`compare.md`](compare.md) | Side-by-side comparison of multiple offers |
+| [`contact.md`](contact.md) | LinkedIn or email outreach drafts |
+| [`deep.md`](deep.md) | Deeper company / role research |
+| [`pdf.md`](pdf.md) | Tailored CV and PDF generation |
+| [`training.md`](training.md) | Evaluate a course, cert, or learning path |
+| [`project.md`](project.md) | Evaluate a portfolio project for job fit |
+| [`tracker.md`](tracker.md) | Application tracker hygiene and status questions |
+| [`apply.md`](apply.md) | Application forms and long-form answers |
+| [`scan.md`](scan.md) | Portal and job-board scanning |
+| [`pipeline.md`](pipeline.md) | Work through pending URLs in `data/pipeline.md` |
+| [`batch.md`](batch.md) | Batch evaluation workflow and TSV-oriented runs |
+| [`followup.md`](followup.md) | What to follow up on next |
+| [`rejection.md`](rejection.md) | Log or process a rejection |
+| [`negotiation.md`](negotiation.md) | Offer received — negotiation framing |
+
+To tailor profile-driven settings, portals, and templates, see [`docs/CUSTOMIZATION.md`](../docs/CUSTOMIZATION.md).
+
+Contributors: see [`CONTRIBUTING.md`](../CONTRIBUTING.md) for branch workflow and the `npm run verify` gate; prefer one cohesive change per PR (for example a single mode or updates under `_shared.md` only).

package/modes/_shared-calibration.md

@@ -0,0 +1,26 @@
+# Scoring Calibration Anchors
+
+**This file is Read on-demand, NOT loaded into the global `instructions` prefix.** It lives separately from `_shared.md` because its contents churn as the candidate accumulates reports — keeping it out of the cached prefix means updates here don't bust the prompt cache for every unrelated session.
+
+**When to Read this file:** right before assigning a final score during evaluation. Once per evaluation, not per dimension.
+
+---
+
+**Use these reference profiles to anchor scores and prevent drift over time.** When evaluating an offer, mentally compare it to these anchors before assigning a final score. Scores MUST be relative to the candidate's actual profile, not absolute.
+
+<!-- [CUSTOMIZE] Replace these with real offers you've evaluated, or archetypes
+     that represent clear score levels for YOUR situation. The examples below
+     are generic starting points — after 10-20 evaluations, replace them with
+     actual reports from your pipeline (e.g., "Report #045 — Anthropic — 4.7/5"). -->
+
+| Score | What it looks like | Example anchor |
+|-------|--------------------|----------------|
+| **5.0** | Dream role. Exact archetype, 90%+ CV match, top-quartile comp, remote, strong brand, fast process. You'd accept immediately. | _Replace with your highest-scored report once you have one_ |
+| **4.0** | Strong match. Right archetype, 75%+ match, fair comp, minor gaps that are easy to frame. Worth a tailored application. | _Replace with a real ~4.0 report_ |
+| **3.0** | Moderate match. Adjacent archetype, 50-60% match, 2-4 hard gaps, comp unknown or median. Worth evaluating but not a priority. | _Replace with a real ~3.0 report_ |
+| **2.0** | Weak match. Wrong seniority or archetype, major gaps, below-market comp signals. Discourage unless specific reason. | _Replace with a real ~2.0 report_ |
+| **1.0** | No fit. Unrelated domain, entry-level, relocation-only, or red flags. Skip. | _Replace with a real ~1.0 report_ |
+
+**Recalibration trigger:** After every 50 evaluations (or when you notice scores clustering — e.g., everything is 3.5-4.2), review the anchors table. Replace generic descriptions with actual reports. If your best offer so far is a 4.3, that's your effective ceiling — adjust the 5.0 anchor to reflect what a true dream role would actually look like for you.
+
+**How to use during evaluation:** After computing the weighted score, sanity-check it against the anchors. "Is this really a 4.5? Is it as strong as [anchor report]?" If not, adjust. The anchors prevent both inflation (everything is 4+) and deflation (nothing breaks 3.5).