Review: Consolidated pipeline optimizations, UI tweaks, and test patches

batch/batch-runner.sh

@@ -27,6 +27,7 @@ DRY_RUN=false
 RETRY_FAILED=false
 START_FROM=0
 MAX_RETRIES=2
+MIN_SCORE=0
 
 usage() {
   cat <<'USAGE'
@@ -41,6 +42,7 @@ Options:
   --retry-failed       Only retry offers marked as "failed" in state
   --start-from N       Start from offer ID N (skip earlier IDs)
   --max-retries N      Max retry attempts per offer (default: 2)
+  --min-score N        Minimum score required to save the report (default: 0)
   -h, --help           Show this help
 
 Files:
@@ -73,6 +75,7 @@ while [[ $# -gt 0 ]]; do
     --retry-failed) RETRY_FAILED=true; shift ;;
     --start-from) START_FROM="$2"; shift 2 ;;
     --max-retries) MAX_RETRIES="$2"; shift 2 ;;
+    --min-score) MIN_SCORE="$2"; shift 2 ;;
     -h|--help) usage; exit 0 ;;
     *) echo "Unknown option: $1"; usage; exit 1 ;;
   esac
@@ -334,6 +337,7 @@ process_offer() {
   local esc_url esc_jd_file esc_report_num esc_date esc_id
   esc_url="${url//\\/\\\\}"
   esc_url="${esc_url//|/\\|}"
+  esc_url="${esc_url//&/\\&}"
   esc_jd_file="${jd_file//\\/\\\\}"
   esc_jd_file="${esc_jd_file//|/\\|}"
   esc_report_num="${report_num//|/\\|}"
@@ -365,13 +369,22 @@ process_offer() {
     # Try to extract score from worker output
     local score="-"
     local score_match
-    score_match=$(grep -oP '"score":\s*[\d.]+' "$log_file" 2>/dev/null | head -1 | grep -oP '[\d.]+' || true)
+    score_match=$(grep -oE '"score":\s*[0-9.]+' "$log_file" 2>/dev/null | head -1 | grep -oE '[0-9.]+' || true)
     if [[ -n "$score_match" ]]; then
       score="$score_match"
     fi
 
-    update_state "$id" "$url" "completed" "$started_at" "$completed_at" "$report_num" "$score" "-" "$retries"
-    echo "    ✅ Completed (score: $score, report: $report_num)"
+    # Check minimum score threshold
+    if [[ "$score" != "-" ]] && awk -v score="$score" -v min="$MIN_SCORE" 'BEGIN{exit !(score >= min)}'; then
+      update_state "$id" "$url" "completed" "$started_at" "$completed_at" "$report_num" "$score" "-" "$retries"
+      echo "    ✅ Completed (score: $score, report: $report_num)"
+    elif [[ "$score" != "-" ]]; then
+      update_state "$id" "$url" "skipped" "$started_at" "$completed_at" "$report_num" "$score" "Score $score below min $MIN_SCORE" "$retries"
+      echo "    ⏭️  Skipped (score: $score < min $MIN_SCORE, report: $report_num)"
+    else
+      update_state "$id" "$url" "completed" "$started_at" "$completed_at" "$report_num" "$score" "-" "$retries"
+      echo "    ✅ Completed (score: -, report: $report_num)"
+    fi
   else
     retries=$((retries + 1))
     local error_msg
@@ -401,7 +414,7 @@ print_summary() {
     return
   fi
 
-  local total=0 completed=0 failed=0 pending=0
+  local total=0 completed=0 failed=0 pending=0 skipped=0
   local score_sum=0 score_count=0
 
   while IFS=$'\t' read -r sid _ sstatus _ _ _ sscore _ _; do
@@ -415,11 +428,12 @@ print_summary() {
         fi
         ;;
       failed) failed=$((failed + 1)) ;;
+      skipped) skipped=$((skipped + 1)) ;;
       *) pending=$((pending + 1)) ;;
     esac
   done < "$STATE_FILE"
 
-  echo "Total: $total | Completed: $completed | Failed: $failed | Pending: $pending"
+  echo "Total: $total | Completed: $completed | Skipped: $skipped | Failed: $failed | Pending: $pending"
 
   if (( score_count > 0 )); then
     local avg

check-liveness.mjs

@@ -17,8 +17,14 @@
 import { chromium } from 'playwright';
 import { readFile } from 'fs/promises';
 import { classifyLiveness } from './liveness-core.mjs';
+import { resolve, dirname } from 'path';
+import { fileURLToPath } from 'url';
 
 async function checkUrl(page, url) {
+  if (!url.startsWith('http://') && !url.startsWith('https://')) {
+    return { result: 'expired', reason: 'Invalid URL protocol' };
+  }
+
   try {
     const response = await page.goto(url, { waitUntil: 'domcontentloaded', timeout: 15000 });
 
@@ -80,7 +86,13 @@ async function main() {
 
   let urls;
   if (args[0] === '--file') {
-    const text = await readFile(args[1], 'utf-8');
+    const filePath = resolve(args[1]);
+    const rootDir = dirname(fileURLToPath(import.meta.url));
+    if (!filePath.startsWith(rootDir)) {
+      console.error('Security Error: Path traversal attempt blocked for --file');
+      process.exit(1);
+    }
+    const text = await readFile(filePath, 'utf-8');
     urls = text.split('\n').map(l => l.trim()).filter(l => l && !l.startsWith('#'));
   } else {
     urls = args;

dashboard/internal/ui/screens/viewer.go

@@ -23,15 +23,86 @@ type ViewerModel struct {
 	theme        theme.Theme
 }
 
+// formatTables aligns markdown tables properly
+func formatTables(lines []string) []string {
+	var out []string
+	var tableLines []string
+
+	flushTable := func() {
+		if len(tableLines) == 0 {
+			return
+		}
+		var colWidths []int
+		var rows [][]string
+		for _, tl := range tableLines {
+			cols := strings.Split(tl, "|")
+			rows = append(rows, cols)
+			for i, c := range cols {
+				// Don't count separator lines for max width
+				if strings.Contains(c, "---") {
+					continue
+				}
+				w := len([]rune(strings.TrimSpace(c)))
+				if i >= len(colWidths) {
+					colWidths = append(colWidths, w)
+				} else if w > colWidths[i] {
+					colWidths[i] = w
+				}
+			}
+		}
+		for _, cols := range rows {
+			var formattedCols []string
+			for i, c := range cols {
+				if i == 0 || i == len(cols)-1 {
+					if strings.TrimSpace(c) == "" {
+						formattedCols = append(formattedCols, "")
+						continue
+					}
+				}
+				w := 0
+				if i < len(colWidths) {
+					w = colWidths[i]
+				}
+				c = strings.TrimSpace(c)
+				if strings.Contains(c, "---") {
+					if w < 3 { w = 3 }
+					formattedCols = append(formattedCols, strings.Repeat("─", w+2))
+				} else {
+					pad := w - len([]rune(c))
+					if pad < 0 { pad = 0 }
+					formattedCols = append(formattedCols, " "+c+strings.Repeat(" ", pad)+" ")
+				}
+			}
+			out = append(out, strings.Join(formattedCols, "|"))
+		}
+		tableLines = nil
+	}
+
+	for _, line := range lines {
+		trimmed := strings.TrimSpace(line)
+		if strings.HasPrefix(trimmed, "|") && strings.HasSuffix(trimmed, "|") {
+			tableLines = append(tableLines, trimmed)
+		} else {
+			flushTable()
+			out = append(out, line)
+		}
+	}
+	flushTable()
+	return out
+}
+
 // NewViewerModel creates a new file viewer for the given path.
 func NewViewerModel(t theme.Theme, path, title string, width, height int) ViewerModel {
 	content, err := os.ReadFile(path)
 	if err != nil {
 		content = []byte("Error reading file: " + err.Error())
 	}
 
+	lines := strings.Split(string(content), "\n")
+	lines = formatTables(lines)
+
 	return ViewerModel{
-		lines:  strings.Split(string(content), "\n"),
+		lines:  lines,
 		title:  title,
 		width:  width,
 		height: height,
@@ -257,7 +328,7 @@ func (m ViewerModel) styleLine(line string) string {
 			Render(line)
 	}
 	// Table headers/separators
-	if strings.HasPrefix(trimmed, "|") && strings.Contains(trimmed, "---") {
+	if strings.HasPrefix(trimmed, "|") && (strings.Contains(trimmed, "---") || strings.Contains(trimmed, "─")) {
 		return lipgloss.NewStyle().
 			Foreground(m.theme.Overlay).
 			Render(line)

dedup-tracker.mjs

@@ -62,11 +62,26 @@ function normalizeRole(role) {
     .trim();
 }
 
+const STOPWORDS = new Set([
+  'senior', 'junior', 'lead', 'manager', 'director', 'intern', 'remote', 
+  'hybrid', 'onsite', 'engineer', 'developer', 'staff', 'principal',
+  'london', 'york', 'francisco', 'seattle', 'boston', 'austin', 'berlin',
+  'paris', 'sydney', 'toronto', 'chicago', 'usa', 'uk', 'eu',
+  'level', 'tier', 'parttime', 'fulltime', 'contract', 'role', 'team'
+]);
+
 function roleMatch(a, b) {
-  const wordsA = normalizeRole(a).split(/\s+/).filter(w => w.length > 3);
-  const wordsB = normalizeRole(b).split(/\s+/).filter(w => w.length > 3);
+  const wordsA = normalizeRole(a).split(/\s+/).filter(w => w.length > 3 && !STOPWORDS.has(w));
+  const wordsB = normalizeRole(b).split(/\s+/).filter(w => w.length > 3 && !STOPWORDS.has(w));
+
+  if (wordsA.length === 0 || wordsB.length === 0) return false;
+
   const overlap = wordsA.filter(w => wordsB.some(wb => wb.includes(w) || w.includes(wb)));
-  return overlap.length >= 2;
+
+  const ratioA = overlap.length / wordsA.length;
+  const ratioB = overlap.length / wordsB.length;
+
+  return overlap.length >= 2 || (overlap.length >= 1 && Math.max(ratioA, ratioB) >= 0.5);
 }
 
 function parseScore(s) {

docs/CUSTOMIZATION.md

@@ -11,9 +11,9 @@ Key sections:
 - **compensation**: Target range, minimum, currency
 - **location**: Country, timezone, visa status, on-site availability
 
-## Target Roles (modes/_shared.md)
+## Target Roles (modes/_profile.md)
 
-The archetype table in `_shared.md` determines how offers are scored and CVs are framed. Edit the table to match YOUR career targets:
+The archetype table in `_profile.md` determines how offers are scored and CVs are framed. Edit the table to match YOUR career targets:
 
 ```markdown
 | Archetype | Thematic axes | What they buy |

generate-pdf.mjs

@@ -91,8 +91,17 @@ async function generatePDF() {
     process.exit(1);
   }
 
-  inputPath = resolve(inputPath);
-  outputPath = resolve(outputPath);
+  function enforceSafePath(p) {
+    const resolvedPath = resolve(p);
+    if (!resolvedPath.startsWith(__dirname)) {
+      console.error(`Security Error: Path traversal attempt blocked for path: ${p}`);
+      process.exit(1);
+    }
+    return resolvedPath;
+  }
+
+  inputPath = enforceSafePath(inputPath);
+  outputPath = enforceSafePath(outputPath);
 
   // Validate format
   const validFormats = ['a4', 'letter'];