ci: validate SKILL.md frontmatter and README skill table (#2)

- Add .github/workflows/validate-skills.yml (runs on PRs + pushes to main)
- Add validator: every skills/*/SKILL.md needs valid YAML frontmatter
  with non-empty name + description; README table must match skills/ dirs
- Fix geo-query-finder/SKILL.md: add missing frontmatter
- Add geo-query-finder + stripe-dispute rows to README table
- Remove stale marketing-psychology row (no skill dir)
- Bump skill count to 65+ in README and USAGE

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

.github/scripts/validate_skills.py

@@ -0,0 +1,83 @@
+#!/usr/bin/env python3
+"""Validate skill frontmatter and README skill table.
+
+Checks:
+1. Every skills/*/SKILL.md has a YAML frontmatter block (delimited by ---).
+2. The frontmatter parses as valid YAML and has non-empty `name` and `description`.
+3. The README skill table lists exactly the skills present in skills/ (count + names).
+
+Exits non-zero with a report if any check fails.
+"""
+import pathlib
+import re
+import sys
+
+import yaml
+
+ROOT = pathlib.Path(__file__).resolve().parents[2]
+SKILLS_DIR = ROOT / "skills"
+README = ROOT / "README.md"
+
+# Matches README table rows like: | `skill-name` | description | ... |
+TABLE_ROW = re.compile(r"^\|\s*`([a-z0-9-]+)`\s*\|")
+
+
+def parse_frontmatter(text: str):
+    """Return the parsed frontmatter dict, or raise ValueError."""
+    if not text.startswith("---"):
+        raise ValueError("missing YAML frontmatter (file must start with '---')")
+    parts = text.split("---", 2)
+    if len(parts) < 3:
+        raise ValueError("frontmatter block is not closed with a second '---'")
+    data = yaml.safe_load(parts[1])
+    if not isinstance(data, dict):
+        raise ValueError("frontmatter is not a YAML mapping")
+    return data
+
+
+def main() -> int:
+    errors = []
+
+    skill_dirs = sorted(p.name for p in SKILLS_DIR.iterdir() if p.is_dir())
+
+    for name in skill_dirs:
+        skill_md = SKILLS_DIR / name / "SKILL.md"
+        if not skill_md.exists():
+            errors.append(f"{name}: missing SKILL.md")
+            continue
+        try:
+            data = parse_frontmatter(skill_md.read_text(encoding="utf-8"))
+        except (ValueError, yaml.YAMLError) as exc:
+            errors.append(f"{name}/SKILL.md: {exc}")
+            continue
+        for field in ("name", "description"):
+            value = data.get(field)
+            if not (isinstance(value, str) and value.strip()):
+                errors.append(f"{name}/SKILL.md: missing or empty `{field}` field")
+
+    # Reconcile the README skill table against the skills/ directory.
+    documented = {m.group(1) for line in README.read_text(encoding="utf-8").splitlines()
+                  if (m := TABLE_ROW.match(line))}
+    actual = set(skill_dirs)
+
+    undocumented = sorted(actual - documented)
+    phantom = sorted(documented - actual)
+    if undocumented:
+        errors.append("README skill table is missing rows for: " + ", ".join(undocumented))
+    if phantom:
+        errors.append("README skill table has rows with no skills/ dir: " + ", ".join(phantom))
+    if len(documented) != len(actual):
+        errors.append(f"README skill count ({len(documented)}) != skills/ dir count ({len(actual)})")
+
+    if errors:
+        print("Skill validation FAILED:\n")
+        for e in errors:
+            print(f"  - {e}")
+        return 1
+
+    print(f"All {len(actual)} skills valid; README table matches skills/ directory.")
+    return 0
+
+
+if __name__ == "__main__":
+    sys.exit(main())

.github/workflows/validate-skills.yml

@@ -0,0 +1,20 @@
+name: Validate Skills
+
+on:
+  push:
+    branches: [main]
+  pull_request:
+    branches: [main]
+
+jobs:
+  validate:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+      - uses: actions/setup-python@v5
+        with:
+          python-version: "3.12"
+      - name: Install PyYAML
+        run: pip install pyyaml
+      - name: Validate SKILL.md frontmatter and README table
+        run: python3 .github/scripts/validate_skills.py

README.md

@@ -40,7 +40,7 @@ Most AI marketing tools charge **$50–300/month** for a chat box that gives you
 **Prerequisites:** [Claude Code](https://docs.anthropic.com/en/docs/claude-code) installed and authenticated.
 
 ```bash
-# Install all 63+ marketing skills
+# Install all 65+ marketing skills
 npx openclaudia install --all
 
 # Or install specific skills
@@ -94,6 +94,7 @@ cp -r skills/seo-audit .claude/skills/         # project-level
 | `schema-markup` | Generate and validate Schema.org structured data |
 | `programmatic-seo` | Create SEO-optimized pages at scale |
 | `ahrefs-research` | Ahrefs Python SDK for backlinks, keywords, domain ratings, and traffic |
+| `geo-query-finder` | Find which ChatGPT search queries mention a brand (GEO visibility) |
 
 ### Content Writing
 | Skill | Description |
@@ -149,7 +150,6 @@ cp -r skills/seo-audit .claude/skills/         # project-level
 | Skill | Description |
 |-------|-------------|
 | `marketing-ideas` | 139 proven marketing ideas by category |
-| `marketing-psychology` | 70+ psychological principles for marketing |
 | `competitor-analysis` | Full competitor strategy breakdown |
 | `pricing-strategy` | Pricing page and strategy optimization |
 | `launch-strategy` | Product launch planning and execution |
@@ -184,6 +184,7 @@ cp -r skills/seo-audit .claude/skills/         # project-level
 | `newsletter` | Newsletter growth, engagement, and monetization |
 | `podcast-marketing` | Podcast production, growth, and promotion |
 | `podcast-edit` | Edit podcast audio — trim, remove fillers, normalize loudness |
+| `stripe-dispute` | Fight Stripe disputes and chargebacks with evidence + counter-dispute |
 
 ### CRM & Outreach (API-Powered)
 | Skill | Description | API Required |

USAGE.md

@@ -1,6 +1,6 @@
 # How to Use OpenClaudia
 
-A practical guide to getting the most out of OpenClaudia's 62+ marketing skills in Claude Code.
+A practical guide to getting the most out of OpenClaudia's 65+ marketing skills in Claude Code.
 
 ---
 

skills/geo-query-finder/SKILL.md

@@ -1,3 +1,13 @@
+---
+name: geo-query-finder
+description: >
+  Find which ChatGPT search queries mention a given brand. Tests long-tail
+  queries against ChatGPT's web-search-enabled model and reports which ones
+  surface the brand. Use when the user asks to "find queries for [brand]",
+  "check GEO visibility", "which queries mention [brand]", "geo query finder",
+  "find AI mentions", or "test ChatGPT queries for [brand]".
+---
+
 # GEO Query Finder
 
 Find which ChatGPT search queries mention a given brand. Tests long-tail queries against ChatGPT's web-search-enabled model and reports which ones surface the brand.