@skill-graph/cli 0.5.6

package/scripts/seed-publication-classification.js

@@ -0,0 +1,282 @@
+#!/usr/bin/env node
+'use strict';
+
+/**
+ * seed-publication-classification.js
+ *
+ * One-off seeder: parses skill-graph/docs/_archived/marketplace-publication-priority-2026-05-18.md
+ * (the hand-curated priority doc, now archived) and emits skill-graph/data/publication-classification.json.
+ *
+ * After this runs once the ledger is hand-edited going forward. Re-running is safe (overwrites with
+ * the same content unless the source doc changes).
+ *
+ * Output is consumed by skill-audit-loop/src/build-skill-audit-worklist.js to build the
+ * publicationQueue array in the unified worklist JSON.
+ *
+ * Usage:
+ *   node skill-graph/scripts/seed-publication-classification.js
+ *   node skill-graph/scripts/seed-publication-classification.js --src <path> --dst <path>
+ */
+
+const fs = require('fs');
+const path = require('path');
+
+const REPO_ROOT = path.resolve(__dirname, '../..');
+
+function parseArgs(argv) {
+  const args = {};
+  for (let i = 0; i < argv.length; i++) {
+    const a = argv[i];
+    if (a.startsWith('--')) {
+      const key = a.slice(2);
+      const next = argv[i + 1];
+      if (next && !next.startsWith('--')) {
+        args[key] = next;
+        i++;
+      } else {
+        args[key] = true;
+      }
+    }
+  }
+  return args;
+}
+
+function normSanitize(cell) {
+  if (!cell) return null;
+  const c = cell.toLowerCase().trim();
+  if (c === 'no') return 'no';
+  if (c === 'minor') return 'minor';
+  if (c === 'light') return 'minor';
+  if (c === 'yes' || c.startsWith('yes')) return 'yes';
+  if (c === '—' || c === '-' || c === '') return null;
+  return c;
+}
+
+function normSource(cell) {
+  if (!cell) return null;
+  const c = cell.toLowerCase().trim();
+  if (c.startsWith('port+sanitize')) return 'port+sanitize';
+  if (c.startsWith('port')) return 'port';
+  if (c.startsWith('rewrite')) return 'rewrite';
+  if (c.startsWith('author')) return 'author';
+  if (c === '—' || c === '-' || c === '') return null;
+  return c;
+}
+
+function normDemand(cell, tier) {
+  if (cell) {
+    const c = cell.toLowerCase();
+    if (c.includes('anthropic') && c.includes('official')) return 'anthropic-official';
+    if (c.includes('skills.sh') || c.includes('leaderboard') || c.includes('installs')) return 'skills.sh-leaderboard';
+    if (c.includes('demand') || c.includes('topic') || c.includes('cluster')) return 'demand-cluster';
+  }
+  if (tier === 'S') return 'skills.sh-leaderboard';
+  if (tier === 'A') return 'demand-cluster';
+  return 'niche';
+}
+
+function extractSkill(cell) {
+  const m = cell.match(/`([a-z0-9][a-z0-9._-]*)`/i);
+  return m ? m[1] : null;
+}
+
+function extractRename(cell) {
+  const m = cell.match(/→\s*([a-z0-9][a-z0-9._-]*)/i);
+  return m ? m[1] : null;
+}
+
+function main() {
+  const args = parseArgs(process.argv.slice(2));
+  const src = args.src
+    ? path.resolve(args.src)
+    : path.join(REPO_ROOT, 'skill-graph/docs/_archived/marketplace-publication-priority-2026-05-18.md');
+  const fallbackSrc = path.join(REPO_ROOT, 'skill-graph/docs/marketplace-publication-priority-2026-05-18.md');
+  const dst = args.dst
+    ? path.resolve(args.dst)
+    : path.join(REPO_ROOT, 'skill-graph/data/publication-classification.json');
+
+  let srcPath = src;
+  if (!fs.existsSync(srcPath) && fs.existsSync(fallbackSrc)) {
+    srcPath = fallbackSrc;
+  }
+  if (!fs.existsSync(srcPath)) {
+    console.error(`Source not found: tried ${src} and ${fallbackSrc}`);
+    process.exit(1);
+  }
+
+  const md = fs.readFileSync(srcPath, 'utf8');
+
+  const ledger = {
+    schema_version: '1.0.0',
+    last_updated: new Date().toISOString(),
+    source_of_classifications: path.relative(REPO_ROOT, srcPath),
+    description: 'Per-skill publication classification + market-state attributes for the OSS marketplace surface. Seeded once from the 2026-05-18 priority doc; hand-edited going forward. Consumed by skill-audit-loop/src/build-skill-audit-worklist.js to build publicationQueue in the unified worklist JSON.',
+    classifications: {
+      publishable: 'Generic / portable; recommended for jacob-balslev/skills publication.',
+      'sales-hub-bound': 'References internal schemas, Printify shop IDs, order_events, customer data, etc. Not publishable.',
+      'personal-infra': "Specific to ~/Development monorepo orchestration; no audience outside this repo.",
+    },
+    skills: {},
+  };
+
+  const lines = md.split('\n');
+  let currentTier = null;
+  let inTable = false;
+  let colIdx = {};
+
+  for (let i = 0; i < lines.length; i++) {
+    const line = lines[i];
+
+    const tierMatch = line.match(/^##\s+Tier\s+([SABC])\b/);
+    if (tierMatch) {
+      currentTier = tierMatch[1];
+      inTable = false;
+      colIdx = {};
+      continue;
+    }
+
+    if (/^##\s+Excluded/.test(line) || /^##\s+Recommended/.test(line) || /^##\s+Completeness/.test(line)) {
+      currentTier = null;
+      inTable = false;
+      continue;
+    }
+
+    if (!currentTier) continue;
+
+    if (/^\|\s*#\s*\|\s*Skill\s*\|/i.test(line)) {
+      const headerCols = line.split('|').map((c) => c.trim()).filter(Boolean);
+      colIdx = {};
+      headerCols.forEach((c, idx) => {
+        const k = c.toLowerCase().replace(/\?.*$/, '').replace(/[^a-z]/g, '');
+        colIdx[k] = idx;
+      });
+      inTable = true;
+      continue;
+    }
+
+    if (inTable && /^\|\s*-+/.test(line)) continue;
+
+    if (inTable && line.startsWith('|') && !/^\|\s*$/.test(line)) {
+      const cells = line.split('|').map((c) => c.trim());
+      const data = cells.slice(1, -1);
+      if (data.length < 3) continue;
+
+      const skillCell = data[colIdx['skill']];
+      if (!skillCell) continue;
+      const skill = extractSkill(skillCell);
+      if (!skill) continue;
+
+      const popCell = data[colIdx['pop']];
+      const popMatch = popCell && popCell.match(/^\s*(\d+)/);
+      const pop_score = popMatch ? parseInt(popMatch[1], 10) : null;
+
+      const source = normSource(data[colIdx['source']]);
+
+      const sanitizeKey = colIdx['sanitize'] !== undefined ? 'sanitize' : null;
+      const needs_sanitization = sanitizeKey ? normSanitize(data[colIdx[sanitizeKey]]) : null;
+
+      const demandKey = colIdx['demandsignal'] !== undefined
+        ? 'demandsignal'
+        : (colIdx['notes'] !== undefined ? 'notes' : null);
+      const demand_signal = demandKey
+        ? normDemand(data[colIdx[demandKey]], currentTier)
+        : normDemand(null, currentTier);
+
+      const notes = (colIdx['notes'] !== undefined ? data[colIdx['notes']] : '') || '';
+      const rename = extractRename(skillCell);
+
+      if (notes && /counted as Tier/i.test(notes) && (!popCell || !popCell.match(/\d/))) continue;
+      if (popCell === '—' || popCell === '-' || popCell === '' || pop_score === null) continue;
+
+      if (!ledger.skills[skill]) {
+        const entry = {
+          classification: 'publishable',
+          tier: currentTier,
+          pop_score,
+        };
+        if (source) entry.source = source;
+        if (needs_sanitization) entry.needs_sanitization = needs_sanitization;
+        if (demand_signal) entry.demand_signal = demand_signal;
+        if (rename) entry.rename_to = rename;
+        if (notes) entry.notes = notes;
+        ledger.skills[skill] = entry;
+      }
+    }
+  }
+
+  // Inline backtick tokens that appear in the SH-bound paragraph but are NOT skills —
+  // they're schema/column names referenced as exclusion reasons.
+  const NON_SKILL_BACKTICK_DENYLIST = new Set([
+    'order_events',
+    'orgQuery',
+    'org_id',
+    'requireOrgAuth',
+    'app_role',
+    'orgScopedQuery',
+    'printify_shop_id',
+    'shop_id',
+  ]);
+
+  const shBoundMatch = md.match(/##\s+Excluded\s+—\s+Sales-Hub-bound[^\n]*\n([\s\S]*?)(?=^##\s)/m);
+  if (shBoundMatch) {
+    const text = shBoundMatch[1];
+    const regex = /`([a-z0-9][a-z0-9._-]*)`(?:\s*\(([^)]+)\))?/gi;
+    let m;
+    while ((m = regex.exec(text)) !== null) {
+      const [, skill, reason] = m;
+      if (NON_SKILL_BACKTICK_DENYLIST.has(skill)) continue;
+      if (!ledger.skills[skill]) {
+        ledger.skills[skill] = {
+          classification: 'sales-hub-bound',
+          tier: null,
+          pop_score: null,
+          notes: reason || 'References Sales-Hub-internal schemas/flows',
+        };
+      }
+    }
+  }
+
+  const piMatch = md.match(/##\s+Excluded\s+—\s+Personal-infra[^\n]*\n([\s\S]*?)(?=^##\s|^---)/m);
+  if (piMatch) {
+    const text = piMatch[1];
+    const regex = /`([a-z0-9][a-z0-9._-]*)`(?:\s*\(([^)]+)\))?/gi;
+    let m;
+    while ((m = regex.exec(text)) !== null) {
+      const [, skill, reason] = m;
+      if (!ledger.skills[skill]) {
+        ledger.skills[skill] = {
+          classification: 'personal-infra',
+          tier: null,
+          pop_score: null,
+          notes: reason || 'Internal to ~/Development monorepo orchestration',
+        };
+      }
+    }
+  }
+
+  const tierCounts = { S: 0, A: 0, B: 0, C: 0 };
+  const classCounts = { publishable: 0, 'sales-hub-bound': 0, 'personal-infra': 0 };
+  for (const entry of Object.values(ledger.skills)) {
+    if (entry.tier) tierCounts[entry.tier]++;
+    classCounts[entry.classification] = (classCounts[entry.classification] || 0) + 1;
+  }
+
+  fs.mkdirSync(path.dirname(dst), { recursive: true });
+  fs.writeFileSync(dst, JSON.stringify(ledger, null, 2) + '\n');
+
+  console.log(`Wrote ${path.relative(REPO_ROOT, dst)}`);
+  console.log(`  Total skills: ${Object.keys(ledger.skills).length}`);
+  console.log(`  Tier counts: S=${tierCounts.S} A=${tierCounts.A} B=${tierCounts.B} C=${tierCounts.C}`);
+  console.log(`  Class counts:`, classCounts);
+}
+
+if (require.main === module) {
+  try {
+    main();
+  } catch (e) {
+    console.error(e.stack || e.message);
+    process.exit(1);
+  }
+}
+
+module.exports = { main };