diff --git a/.github/workflows/pr-check.yml b/.github/workflows/pr-check.yml
index 4e56533..d30babe 100644
--- a/.github/workflows/pr-check.yml
+++ b/.github/workflows/pr-check.yml
@@ -4,86 +4,76 @@ on:
   pull_request:
     branches: [main]
     types: [opened, edited, synchronize, reopened]
+    paths:
+      - 'awesome-privacy.yml'
+      - '.github/README.md'
 
 permissions:
   contents: read
   pull-requests: read
 
 jobs:
-  detect-changes:
-    name: Detect changes
+  pr-compliance:
+    name: PR Compliance
     runs-on: ubuntu-latest
-    outputs:
-      yaml_changed: ${{ steps.changes.outputs.yaml_changed }}
-      non_yaml_changed: ${{ steps.changes.outputs.non_yaml_changed }}
     steps:
       - uses: actions/checkout@v4
       - run: git fetch --depth=1 origin ${{ github.event.pull_request.base.sha }}
       - uses: actions/setup-python@v5
         with:
           python-version: "3.12"
-      - name: Detect changed files
-        id: changes
-        run: python lib/checks/detect-changes.py --base-ref ${{ github.event.pull_request.base.sha }}
-      - name: Non-YAML changes warning
-        if: steps.changes.outputs.non_yaml_changed == 'true'
-        run: python lib/checks/warn-non-yaml.py --base-ref ${{ github.event.pull_request.base.sha }}
-
-  pr-meta:
-    name: PR metadata
-    runs-on: ubuntu-latest
-    steps:
-      - uses: actions/checkout@v4
-      - uses: actions/setup-python@v5
-        with:
-          python-version: "3.12"
+      - name: Check README edits
+        id: readme
+        continue-on-error: true
+        run: python lib/checks/check-readme-edits.py --base-ref ${{ github.event.pull_request.base.sha }}
       - name: Check PR metadata
+        id: meta
         env:
           PR_TITLE: ${{ github.event.pull_request.title }}
           PR_BODY: ${{ github.event.pull_request.body }}
           PR_DRAFT: ${{ github.event.pull_request.draft }}
+          README_FAILED: ${{ steps.readme.outcome == 'failure' && 'true' || 'false' }}
         run: python lib/checks/check-pr-meta.py
       - name: Upload findings
         if: always()
         uses: actions/upload-artifact@v4
         with:
-          name: findings-meta
-          path: /tmp/findings-meta.json
+          name: findings-compliance
+          path: /tmp/findings-compliance.json
           if-no-files-found: ignore
-
-  file-checks:
-    name: File checks
-    runs-on: ubuntu-latest
-    steps:
-      - uses: actions/checkout@v4
-      - run: git fetch --depth=1 origin ${{ github.event.pull_request.base.sha }}
-      - uses: actions/setup-python@v5
-        with:
-          python-version: "3.12"
-      - name: Check for direct README edits
-        run: python lib/checks/check-readme-edits.py --base-ref ${{ github.event.pull_request.base.sha }}
+      - name: Fail if critical
+        if: steps.readme.outcome == 'failure' || steps.meta.outcome == 'failure'
+        run: exit 1
 
   data-validation:
-    name: Data validation
-    needs: detect-changes
-    if: needs.detect-changes.outputs.yaml_changed == 'true'
+    name: Data Validation
     runs-on: ubuntu-latest
+    outputs:
+      yaml_changed: ${{ steps.changes.outputs.yaml_changed }}
     steps:
       - uses: actions/checkout@v4
       - run: git fetch --depth=1 origin ${{ github.event.pull_request.base.sha }}
       - uses: actions/setup-python@v5
         with:
           python-version: "3.12"
-      - run: pip install -q -r lib/requirements.txt
+      - name: Detect changes
+        id: changes
+        run: python lib/checks/detect-changes.py --base-ref ${{ github.event.pull_request.base.sha }}
+      - name: Install dependencies
+        if: steps.changes.outputs.yaml_changed == 'true'
+        run: pip install -q -r lib/requirements.txt
       - name: Schema validation
+        if: steps.changes.outputs.yaml_changed == 'true'
         id: schema
         continue-on-error: true
         run: make validate
       - name: YAML diff
+        if: steps.changes.outputs.yaml_changed == 'true'
         id: diff
         continue-on-error: true
         run: python lib/checks/check-yaml-diff.py --base-ref ${{ github.event.pull_request.base.sha }}
       - name: Check additions
+        if: steps.changes.outputs.yaml_changed == 'true'
         env:
           SCHEMA_OUTCOME: ${{ steps.schema.outcome }}
         run: python lib/checks/check-additions.py
@@ -101,14 +91,14 @@ jobs:
           name: findings-data
           path: /tmp/findings-data.json
           if-no-files-found: ignore
-      - name: Fail if critical checks failed
-        if: steps.schema.outcome == 'failure' || steps.diff.outcome == 'failure'
+      - name: Fail if critical
+        if: steps.changes.outputs.yaml_changed == 'true' && (steps.schema.outcome == 'failure' || steps.diff.outcome == 'failure')
         run: exit 1
 
-  project-checks:
-    name: Project checks
+  submission-eligibility:
+    name: Submission Eligibility
     needs: data-validation
-    if: "!cancelled() && needs.data-validation.result != 'skipped'"
+    if: "!cancelled() && needs.data-validation.outputs.yaml_changed == 'true'"
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v4
@@ -138,7 +128,7 @@ jobs:
   summary:
     name: Summary
     if: always()
-    needs: [detect-changes, pr-meta, file-checks, data-validation, project-checks]
+    needs: [pr-compliance, data-validation, submission-eligibility]
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v4
@@ -157,7 +147,6 @@ jobs:
           PR_USER: ${{ github.event.pull_request.user.login }}
           PR_NUMBER: ${{ github.event.pull_request.number }}
           RUN_ID: ${{ github.run_id }}
-          README_FAILED: ${{ needs.file-checks.result == 'failure' && 'true' || 'false' }}
         run: python lib/checks/format-comment.py
       - name: Upload PR metadata
         if: always()
diff --git a/lib/checks/check-links.py b/lib/checks/check-links.py
deleted file mode 100644
index 293fa24..0000000
--- a/lib/checks/check-links.py
+++ /dev/null
@@ -1,211 +0,0 @@
-"""
-Validates URLs and metadata for added/modified services in a PR.
-Reads the diff JSON produced by check-yaml-diff.py.
-All checks are warnings only -- this script never fails (exit 0).
-"""
-
-import argparse
-import json
-import os
-import sys
-
-import requests
-import yaml
-
-# Paths (relative to project root)
-PROJECT_ROOT = os.path.dirname(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
-DATA_PATH = os.path.join(PROJECT_ROOT, "awesome-privacy.yml")
-
-# Exit codes
-EXIT_PASS = 0
-EXIT_RUNTIME_ERROR = 2
-
-# ANSI color helpers
-_use_color = sys.stderr.isatty() and not os.environ.get("NO_COLOR")
-red = (lambda s: f"\033[31m{s}\033[0m") if _use_color else (lambda s: s)
-green = (lambda s: f"\033[32m{s}\033[0m") if _use_color else (lambda s: s)
-yellow = (lambda s: f"\033[33m{s}\033[0m") if _use_color else (lambda s: s)
-
-TIMEOUT = 10
-USER_AGENT = "awesome-privacy-ci/1.0"
-DESC_MIN_LEN = 50
-DESC_MAX_LEN = 250
-
-
-def check_url(url, label):
-    """Check if a URL is reachable. Returns (ok, message)."""
-    try:
-        resp = requests.head(
-            url,
-            timeout=TIMEOUT,
-            allow_redirects=True,
-            headers={"User-Agent": USER_AGENT},
-        )
-        if resp.status_code >= 400:
-            # Retry with GET -- some servers reject HEAD
-            resp = requests.get(
-                url,
-                timeout=TIMEOUT,
-                allow_redirects=True,
-                headers={"User-Agent": USER_AGENT},
-                stream=True,
-            )
-            resp.close()
-        if resp.status_code >= 400:
-            return False, f"{label}: HTTP {resp.status_code} for {url}"
-        return True, None
-    except requests.RequestException as e:
-        return False, f"{label}: Connection error for {url} ({type(e).__name__})"
-
-
-def check_service(service_data, service_name, category, section):
-    """Run all checks on a single service. Returns list of warning strings."""
-    warnings = []
-    name_prefix = f"{category} > {section} > {service_name}"
-
-    # Check url
-    url = service_data.get("url")
-    if url:
-        ok, msg = check_url(url, "url")
-        if not ok:
-            warnings.append(f"{name_prefix}: {msg}")
-    else:
-        warnings.append(f"{name_prefix}: missing required field 'url'")
-
-    # Check icon
-    icon = service_data.get("icon")
-    if icon:
-        ok, msg = check_url(icon, "icon")
-        if not ok:
-            warnings.append(f"{name_prefix}: {msg}")
-    else:
-        warnings.append(f"{name_prefix}: missing 'icon' field (recommended by contributing guide)")
-
-    # Check iosApp
-    ios_app = service_data.get("iosApp")
-    if ios_app:
-        ok, msg = check_url(ios_app, "iosApp")
-        if not ok:
-            warnings.append(f"{name_prefix}: {msg}")
-
-    # Check github
-    github = service_data.get("github")
-    if github:
-        github_url = f"https://github.com/{github}"
-        ok, msg = check_url(github_url, "github")
-        if not ok:
-            warnings.append(f"{name_prefix}: {msg}")
-
-    # Check description length
-    desc = service_data.get("description", "")
-    desc_stripped = desc.strip()
-    desc_len = len(desc_stripped)
-    if desc_len < DESC_MIN_LEN:
-        warnings.append(f"{name_prefix}: description too short ({desc_len} chars, minimum {DESC_MIN_LEN})")
-    elif desc_len > DESC_MAX_LEN:
-        warnings.append(f"{name_prefix}: description too long ({desc_len} chars, maximum {DESC_MAX_LEN})")
-
-    return warnings
-
-
-def find_service_in_head(category, section, service_name):
-    """Look up a service in the head YAML by category/section/service name."""
-    try:
-        with open(DATA_PATH, "r") as f:
-            data = yaml.safe_load(f)
-        for cat in data.get("categories", []):
-            if cat.get("name") == category:
-                for sec in cat.get("sections", []):
-                    if sec.get("name") == section:
-                        for svc in sec.get("services", []):
-                            if svc.get("name") == service_name:
-                                return svc
-    except Exception:
-        pass
-    return None
-
-
-def write_step_summary(all_warnings, services_checked):
-    """Write a Markdown summary to $GITHUB_STEP_SUMMARY."""
-    summary_file = os.environ.get("GITHUB_STEP_SUMMARY")
-    if not summary_file:
-        return
-
-    lines = ["## Link Validation\n"]
-
-    if not services_checked:
-        lines.append("No services to check.\n")
-    elif not all_warnings:
-        lines.append(f"All checks passed for {services_checked} service(s).\n")
-    else:
-        lines.append(f"Checked {services_checked} service(s), found {len(all_warnings)} warning(s):\n")
-        lines.append("| Warning |")
-        lines.append("|---------|")
-        for w in all_warnings:
-            escaped = w.replace("|", "\\|")
-            lines.append(f"| {escaped} |")
-        lines.append("")
-        lines.append("> **Note:** Link warnings are informational only and do not fail the check. "
-                      "URLs may be temporarily down or block automated requests.\n")
-
-    with open(summary_file, "a") as f:
-        f.write("\n".join(lines) + "\n")
-
-
-def main():
-    parser = argparse.ArgumentParser(description="Validate links for added/modified services")
-    parser.add_argument("--diff-json", required=True, help="Path to the diff JSON file")
-    args = parser.parse_args()
-
-    # Load diff
-    try:
-        with open(args.diff_json, "r") as f:
-            diff = json.load(f)
-    except (FileNotFoundError, json.JSONDecodeError) as e:
-        print(red(f"Failed to load diff JSON: {e}"), file=sys.stderr)
-        sys.exit(EXIT_RUNTIME_ERROR)
-
-    all_warnings = []
-    services_checked = 0
-
-    # Check added services
-    for svc in diff.get("services", {}).get("added", []):
-        services_checked += 1
-        warnings = check_service(
-            svc.get("fields", {}),
-            svc["service"],
-            svc["category"],
-            svc["section"],
-        )
-        all_warnings.extend(warnings)
-
-    # Check modified services -- only if they have URL-related field changes
-    for svc in diff.get("services", {}).get("modified", []):
-        url_fields = {"url", "icon", "iosApp", "github", "description"}
-        changed = set(svc.get("changed_fields", []))
-        if changed & url_fields:
-            services_checked += 1
-            head_svc = find_service_in_head(svc["category"], svc["section"], svc["service"])
-            if head_svc:
-                warnings = check_service(
-                    head_svc,
-                    svc["service"],
-                    svc["category"],
-                    svc["section"],
-                )
-                all_warnings.extend(warnings)
-
-    # Print results
-    if all_warnings:
-        print(yellow(f"Link validation: {len(all_warnings)} warning(s)"), file=sys.stderr)
-        for w in all_warnings:
-            print(f"  {yellow('WARNING')} {w}", file=sys.stderr)
-    else:
-        print(green(f"Link validation passed. {services_checked} service(s) checked."))
-
-    write_step_summary(all_warnings, services_checked)
-    sys.exit(EXIT_PASS)
-
-
-if __name__ == "__main__":
-    main()
diff --git a/lib/checks/check-pr-meta.py b/lib/checks/check-pr-meta.py
index f63f561..aa1379e 100644
--- a/lib/checks/check-pr-meta.py
+++ b/lib/checks/check-pr-meta.py
@@ -5,7 +5,7 @@ import os
 import re
 import sys
 
-FINDINGS_PATH = "/tmp/findings-meta.json"
+FINDINGS_PATH = "/tmp/findings-compliance.json"
 
 BAD_TITLES = {"update readme.md", "update awesome-privacy.yml"}
 
@@ -27,6 +27,11 @@ CHECKBOX_MSG = (
     " to confirm that you've read the contributing guidelines, checked your submission,"
     " indicated your affiliation and agree to follow our CoC"
 )
+README_MSG = (
+    "Do not edit the README directly. This file is auto-generated from the"
+    " content in `awesome-privacy.yml`, and so your changes will be overridden!"
+    " Instead, only modify the YAML file, and be sure to follow our Contributing Guidelines."
+)
 
 
 def extract_section(body, header):
@@ -78,6 +83,13 @@ def check_checkboxes(body):
     return None
 
 
+def check_readme(readme_failed):
+    """Return a finding if the README check reported a failure."""
+    if readme_failed == "true":
+        return README_MSG
+    return None
+
+
 def write_findings(findings):
     """Write the findings list to the output JSON file."""
     with open(FINDINGS_PATH, "w") as f:
@@ -91,6 +103,7 @@ def main():
         title = os.environ.get("PR_TITLE", "")
         body = os.environ.get("PR_BODY", "")
         draft = os.environ.get("PR_DRAFT", "false")
+        readme_failed = os.environ.get("README_FAILED", "false")
 
         finding = check_title(title)
         if finding:
@@ -111,6 +124,10 @@ def main():
             finding = check_checkboxes(body)
             if finding:
                 findings.append(finding)
+
+        finding = check_readme(readme_failed)
+        if finding:
+            findings.append(finding)
     except Exception:
         pass
 
diff --git a/lib/checks/check-template.py b/lib/checks/check-template.py
deleted file mode 100644
index b53671d..0000000
--- a/lib/checks/check-template.py
+++ /dev/null
@@ -1,206 +0,0 @@
-"""
-Checks PR body against the pull request template.
-Reads the PR body from the PR_BODY environment variable (avoids shell injection).
-Exits with code 1 for severe violations (empty body, missing required sections).
-"""
-
-import os
-import re
-import sys
-
-# Exit codes
-EXIT_PASS = 0
-EXIT_FAIL = 1
-EXIT_RUNTIME_ERROR = 2
-
-# Warnings that should cause a hard failure
-CRITICAL_WARNINGS = {
-    "PR body is empty or not provided",
-    "Type section is missing",
-    "Type section is empty",
-    "Changes section is missing",
-    "Changes section is empty",
-    "Checklist section is missing",
-    "Checklist section does not contain any checkbox items",
-}
-
-# ANSI color helpers
-_use_color = sys.stderr.isatty() and not os.environ.get("NO_COLOR")
-red = (lambda s: f"\033[31m{s}\033[0m") if _use_color else (lambda s: s)
-green = (lambda s: f"\033[32m{s}\033[0m") if _use_color else (lambda s: s)
-yellow = (lambda s: f"\033[33m{s}\033[0m") if _use_color else (lambda s: s)
-
-# Valid PR types from the template
-VALID_TYPES = {"Addition", "Amendment", "Removal", "Spelling or Grammar", "Website Update", "Misc"}
-
-# Raw template text that indicates an unfilled section
-RAW_TYPE_LINE = "Addition / Amendment / Removal / Spelling or Grammar / Website Update / Misc"
-
-
-def strip_html_comments(text):
-    """Remove <!-- ... --> comments from text."""
-    return re.sub(r"<!--.*?-->", "", text, flags=re.DOTALL).strip()
-
-
-def extract_section(body, header):
-    """Extract content between a ### header and the next --- or ### header."""
-    pattern = rf"###\s*{re.escape(header)}\s*\n(.*?)(?=\n---|\n###|\Z)"
-    match = re.search(pattern, body, re.DOTALL)
-    if match:
-        return match.group(1)
-    return None
-
-
-def check_type_section(content):
-    """Check the Type section. Returns list of warning strings."""
-    warnings = []
-
-    if content is None:
-        warnings.append("Type section is missing")
-        return warnings
-
-    cleaned = strip_html_comments(content).strip()
-
-    if not cleaned:
-        warnings.append("Type section is empty")
-        return warnings
-
-    # Check for raw template text (unchanged)
-    if RAW_TYPE_LINE in cleaned:
-        warnings.append("Type section appears unchanged from the template -- please select one type")
-        return warnings
-
-    # Check how many valid types are present
-    found_types = [t for t in VALID_TYPES if t in cleaned]
-
-    if len(found_types) == 0:
-        warnings.append(f"Type section does not contain a recognized type. Expected one of: {', '.join(sorted(VALID_TYPES))}")
-    elif len(found_types) > 1:
-        warnings.append(f"Type section contains multiple types ({', '.join(found_types)}) -- please select only one")
-
-    return warnings
-
-
-def check_text_section(content, section_name):
-    """Check a text section (Changes, Supporting Material, Affiliation). Returns list of warnings."""
-    warnings = []
-
-    if content is None:
-        warnings.append(f"{section_name} section is missing")
-        return warnings
-
-    cleaned = strip_html_comments(content).strip()
-
-    if not cleaned:
-        warnings.append(f"{section_name} section is empty")
-
-    return warnings
-
-
-def check_checklist(content):
-    """Check the Checklist section. Returns list of warnings."""
-    warnings = []
-
-    if content is None:
-        warnings.append("Checklist section is missing")
-        return warnings
-
-    checked = re.findall(r"- \[x\]", content, re.IGNORECASE)
-    unchecked = re.findall(r"- \[ \]", content)
-
-    total = len(checked) + len(unchecked)
-
-    if total == 0:
-        warnings.append("Checklist section does not contain any checkbox items")
-        return warnings
-
-    if unchecked:
-        warnings.append(f"Checklist has {len(unchecked)} unchecked item(s) out of {total}")
-
-    return warnings
-
-
-def has_critical_warnings(warnings):
-    """Return True if any warning is a critical (hard-fail) violation."""
-    return any(w in CRITICAL_WARNINGS for w in warnings)
-
-
-def write_step_summary(all_warnings):
-    """Write a Markdown summary to $GITHUB_STEP_SUMMARY."""
-    summary_file = os.environ.get("GITHUB_STEP_SUMMARY")
-    if not summary_file:
-        return
-
-    lines = ["## PR Template Check\n"]
-
-    if not all_warnings:
-        lines.append("All template checks passed.\n")
-    else:
-        critical = has_critical_warnings(all_warnings)
-        lines.append(f"Found {len(all_warnings)} warning(s):\n")
-        for w in all_warnings:
-            lines.append(f"- {w}")
-        lines.append("")
-        if critical:
-            lines.append("> **Error:** One or more required sections are missing or empty. "
-                          "Please fill out the PR template before this check can pass.\n")
-        else:
-            lines.append("> **Note:** Template warnings are informational and do not fail the check. "
-                          "Reviewers will verify compliance.\n")
-
-    with open(summary_file, "a") as f:
-        f.write("\n".join(lines) + "\n")
-
-
-def main():
-    pr_body = os.environ.get("PR_BODY")
-
-    if pr_body is None or pr_body.strip() == "":
-        all_warnings = [
-            "PR body is empty or not provided",
-            "Type section is missing",
-            "Changes section is missing",
-            "Supporting Material section is missing",
-            "Affiliation section is missing",
-            "Checklist section is missing",
-        ]
-        print(yellow(f"PR template check: {len(all_warnings)} warning(s)"), file=sys.stderr)
-        for w in all_warnings:
-            print(f"  {yellow('WARNING')} {w}", file=sys.stderr)
-        write_step_summary(all_warnings)
-        sys.exit(EXIT_FAIL)
-
-    all_warnings = []
-
-    # Check each section
-    type_content = extract_section(pr_body, "Type")
-    all_warnings.extend(check_type_section(type_content))
-
-    changes_content = extract_section(pr_body, "Changes")
-    all_warnings.extend(check_text_section(changes_content, "Changes"))
-
-    supporting_content = extract_section(pr_body, "Supporting Material")
-    all_warnings.extend(check_text_section(supporting_content, "Supporting Material"))
-
-    affiliation_content = extract_section(pr_body, "Affiliation")
-    all_warnings.extend(check_text_section(affiliation_content, "Affiliation"))
-
-    checklist_content = extract_section(pr_body, "Checklist")
-    all_warnings.extend(check_checklist(checklist_content))
-
-    # Print results
-    if all_warnings:
-        print(yellow(f"PR template check: {len(all_warnings)} warning(s)"), file=sys.stderr)
-        for w in all_warnings:
-            print(f"  {yellow('WARNING')} {w}", file=sys.stderr)
-    else:
-        print(green("PR template check passed."))
-
-    write_step_summary(all_warnings)
-    if has_critical_warnings(all_warnings):
-        sys.exit(EXIT_FAIL)
-    sys.exit(EXIT_PASS)
-
-
-if __name__ == "__main__":
-    main()
diff --git a/lib/checks/check-yaml-diff.py b/lib/checks/check-yaml-diff.py
index 2931e68..44a5285 100644
--- a/lib/checks/check-yaml-diff.py
+++ b/lib/checks/check-yaml-diff.py
@@ -1,7 +1,5 @@
-"""
-Analyzes the diff between base and head versions of awesome-privacy.yml.
-Enforces the single-entry rule: only one service addition/amendment/removal per PR.
-Outputs a JSON diff to /tmp/pr-diff.json and writes a step summary.
+"""Analyzes the diff between base and head versions of awesome-privacy.yml.
+Enforces the single-entry rule and outputs a JSON diff to /tmp/pr-diff.json.
 """
 
 import argparse
@@ -12,22 +10,18 @@ import sys
 
 import yaml
 
-# Paths (relative to project root)
 PROJECT_ROOT = os.path.dirname(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 DATA_PATH = os.path.join(PROJECT_ROOT, "awesome-privacy.yml")
 DIFF_OUTPUT_PATH = "/tmp/pr-diff.json"
 
-# Exit codes
 EXIT_PASS = 0
 EXIT_RULE_VIOLATION = 1
 EXIT_RUNTIME_ERROR = 2
 
-# ANSI color helpers
 _use_color = sys.stderr.isatty() and not os.environ.get("NO_COLOR")
 red = (lambda s: f"\033[31m{s}\033[0m") if _use_color else (lambda s: s)
 green = (lambda s: f"\033[32m{s}\033[0m") if _use_color else (lambda s: s)
 yellow = (lambda s: f"\033[33m{s}\033[0m") if _use_color else (lambda s: s)
-dim = (lambda s: f"\033[2m{s}\033[0m") if _use_color else (lambda s: s)
 
 
 def load_base_yaml(base_ref):
@@ -35,13 +29,11 @@ def load_base_yaml(base_ref):
     try:
         result = subprocess.run(
             ["git", "show", f"{base_ref}:awesome-privacy.yml"],
-            capture_output=True, text=True, check=True,
-            cwd=PROJECT_ROOT,
+            capture_output=True, text=True, check=True, cwd=PROJECT_ROOT,
         )
         return yaml.safe_load(result.stdout)
     except subprocess.CalledProcessError:
-        # File doesn't exist in base (completely new file)
-        print(yellow("Warning: awesome-privacy.yml not found in base ref, treating as empty"), file=sys.stderr)
+        print(yellow("awesome-privacy.yml not found in base ref, treating as empty"), file=sys.stderr)
         return {"categories": []}
     except yaml.YAMLError as e:
         print(red(f"Failed to parse base YAML: {e}"), file=sys.stderr)
@@ -51,165 +43,45 @@ def load_base_yaml(base_ref):
 def load_head_yaml():
     """Load the YAML from the current working tree."""
     try:
-        with open(DATA_PATH, "r") as f:
+        with open(DATA_PATH) as f:
             return yaml.safe_load(f)
-    except FileNotFoundError:
-        print(red(f"File not found: {DATA_PATH}"), file=sys.stderr)
-        sys.exit(EXIT_RUNTIME_ERROR)
-    except yaml.YAMLError as e:
-        print(red(f"Failed to parse head YAML: {e}"), file=sys.stderr)
+    except (FileNotFoundError, yaml.YAMLError) as e:
+        print(red(f"Failed to load head YAML: {e}"), file=sys.stderr)
         sys.exit(EXIT_RUNTIME_ERROR)
 
 
-def build_service_index(data):
-    """Build a dict keyed by (category, section, service_name) -> service dict."""
+def build_index(data, depth):
+    """Build a keyed index at the given depth (3=services, 2=sections, 1=categories)."""
     index = {}
     for cat in data.get("categories", []):
-        cat_name = cat.get("name", "")
+        cn = cat.get("name", "")
+        if depth == 1:
+            index[cn] = {k: v for k, v in cat.items() if k != "sections"}
+            continue
         for sec in cat.get("sections", []):
-            sec_name = sec.get("name", "")
+            sn = sec.get("name", "")
+            if depth == 2:
+                index[(cn, sn)] = {k: v for k, v in sec.items() if k != "services"}
+                continue
             for svc in sec.get("services", []):
-                svc_name = svc.get("name", "")
-                key = (cat_name, sec_name, svc_name)
-                index[key] = svc
+                index[(cn, sn, svc.get("name", ""))] = svc
     return index
 
 
-def build_section_index(data):
-    """Build a dict keyed by (category, section) -> section metadata (excluding services)."""
-    index = {}
-    for cat in data.get("categories", []):
-        cat_name = cat.get("name", "")
-        for sec in cat.get("sections", []):
-            sec_name = sec.get("name", "")
-            key = (cat_name, sec_name)
-            meta = {k: v for k, v in sec.items() if k != "services"}
-            index[key] = meta
-    return index
-
-
-def build_category_index(data):
-    """Build a dict keyed by category_name -> category metadata (excluding sections)."""
-    index = {}
-    for cat in data.get("categories", []):
-        cat_name = cat.get("name", "")
-        meta = {k: v for k, v in cat.items() if k != "sections"}
-        index[cat_name] = meta
-    return index
-
-
-def diff_services(base_data, head_data):
-    """Find added, removed, and modified services."""
-    base_idx = build_service_index(base_data)
-    head_idx = build_service_index(head_data)
-
-    base_keys = set(base_idx.keys())
-    head_keys = set(head_idx.keys())
-
-    added = []
-    for key in sorted(head_keys - base_keys):
-        added.append({
-            "category": key[0],
-            "section": key[1],
-            "service": key[2],
-            "fields": head_idx[key],
-        })
-
-    removed = []
-    for key in sorted(base_keys - head_keys):
-        removed.append({
-            "category": key[0],
-            "section": key[1],
-            "service": key[2],
-        })
-
+def diff_index(base_idx, head_idx):
+    """Return (added_keys, removed_keys, modified_keys_with_changed_fields)."""
+    base_keys, head_keys = set(base_idx), set(head_idx)
+    added = sorted(head_keys - base_keys)
+    removed = sorted(base_keys - head_keys)
     modified = []
     for key in sorted(base_keys & head_keys):
-        base_svc = base_idx[key]
-        head_svc = head_idx[key]
-        if base_svc != head_svc:
-            changed_fields = []
-            all_fields = set(base_svc.keys()) | set(head_svc.keys())
-            for field in sorted(all_fields):
-                old_val = base_svc.get(field)
-                new_val = head_svc.get(field)
-                if old_val != new_val:
-                    changed_fields.append(field)
-            modified.append({
-                "category": key[0],
-                "section": key[1],
-                "service": key[2],
-                "changed_fields": changed_fields,
-            })
-
+        if base_idx[key] != head_idx[key]:
+            all_fields = set(base_idx[key]) | set(head_idx[key])
+            changed = sorted(f for f in all_fields if base_idx[key].get(f) != head_idx[key].get(f))
+            modified.append((key, changed))
     return added, removed, modified
 
 
-def diff_sections(base_data, head_data):
-    """Find section-level metadata changes (intro, wordOfWarning, etc.)."""
-    base_idx = build_section_index(base_data)
-    head_idx = build_section_index(head_data)
-
-    base_keys = set(base_idx.keys())
-    head_keys = set(head_idx.keys())
-
-    changes = []
-
-    # New sections
-    for key in sorted(head_keys - base_keys):
-        changes.append({
-            "category": key[0],
-            "section": key[1],
-            "change_type": "added_section",
-        })
-
-    # Removed sections
-    for key in sorted(base_keys - head_keys):
-        changes.append({
-            "category": key[0],
-            "section": key[1],
-            "change_type": "removed_section",
-        })
-
-    # Modified section metadata
-    for key in sorted(base_keys & head_keys):
-        base_meta = base_idx[key]
-        head_meta = head_idx[key]
-        if base_meta != head_meta:
-            changed_fields = []
-            all_fields = set(base_meta.keys()) | set(head_meta.keys())
-            for field in sorted(all_fields):
-                if base_meta.get(field) != head_meta.get(field):
-                    changed_fields.append(field)
-            changes.append({
-                "category": key[0],
-                "section": key[1],
-                "change_type": "modified_section_metadata",
-                "changed_fields": changed_fields,
-            })
-
-    return changes
-
-
-def diff_categories(base_data, head_data):
-    """Find structural category changes."""
-    base_idx = build_category_index(base_data)
-    head_idx = build_category_index(head_data)
-
-    base_keys = set(base_idx.keys())
-    head_keys = set(head_idx.keys())
-
-    changes = []
-
-    for name in sorted(head_keys - base_keys):
-        changes.append({"category": name, "change_type": "added_category"})
-
-    for name in sorted(base_keys - head_keys):
-        changes.append({"category": name, "change_type": "removed_category"})
-
-    return changes
-
-
 def write_github_output(name, value):
     """Write a value to $GITHUB_OUTPUT."""
     output_file = os.environ.get("GITHUB_OUTPUT")
@@ -218,123 +90,113 @@ def write_github_output(name, value):
             f.write(f"{name}={value}\n")
 
 
+def fmt_path(key):
+    """Format a tuple key as a readable path."""
+    return " → ".join(key) if isinstance(key, tuple) else key
+
+
 def write_step_summary(diff_result):
-    """Write a Markdown summary to $GITHUB_STEP_SUMMARY."""
+    """Write a bullet-point Markdown summary to $GITHUB_STEP_SUMMARY."""
     summary_file = os.environ.get("GITHUB_STEP_SUMMARY")
     if not summary_file:
         return
 
     lines = ["## YAML Diff Analysis\n"]
+    bullets = []
 
-    added = diff_result["services"]["added"]
-    removed = diff_result["services"]["removed"]
-    modified = diff_result["services"]["modified"]
-    section_changes = diff_result["sections"]
-    category_changes = diff_result["categories"]
+    for svc in diff_result["services"]["added"]:
+        bullets.append(f"- Added **{svc['service']}** in {svc['category']} → {svc['section']}")
+    for svc in diff_result["services"]["removed"]:
+        bullets.append(f"- Removed **{svc['service']}** from {svc['category']} → {svc['section']}")
+    for svc in diff_result["services"]["modified"]:
+        fields = ", ".join(f"`{f}`" for f in svc["changed_fields"])
+        bullets.append(f"- Modified {fields} in {svc['category']} → {svc['section']} → {svc['service']}")
+    for change in diff_result["sections"]:
+        ct = change["change_type"]
+        path = f"{change['category']} → {change['section']}"
+        if ct == "added_section":
+            bullets.append(f"- Added section **{change['section']}** in {change['category']}")
+        elif ct == "removed_section":
+            bullets.append(f"- Removed section **{change['section']}** from {change['category']}")
+        else:
+            fields = ", ".join(f"`{f}`" for f in change.get("changed_fields", []))
+            bullets.append(f"- Modified section metadata ({fields}) in {path}")
+    for change in diff_result["categories"]:
+        if change["change_type"] == "added_category":
+            bullets.append(f"- Added category **{change['category']}**")
+        else:
+            bullets.append(f"- Removed category **{change['category']}**")
 
-    if not added and not removed and not modified and not section_changes and not category_changes:
-        lines.append("No changes detected in `awesome-privacy.yml`.\n")
+    if bullets:
+        lines.extend(bullets)
     else:
-        lines.append("| Type | Category | Section | Service | Details |")
-        lines.append("|------|----------|---------|---------|---------|")
-
-        for svc in added:
-            lines.append(f"| Added | {svc['category']} | {svc['section']} | {svc['service']} | New service |")
-
-        for svc in removed:
-            lines.append(f"| Removed | {svc['category']} | {svc['section']} | {svc['service']} | Service removed |")
-
-        for svc in modified:
-            fields = ", ".join(svc["changed_fields"])
-            lines.append(f"| Modified | {svc['category']} | {svc['section']} | {svc['service']} | Changed: {fields} |")
-
-        for change in section_changes:
-            detail = change["change_type"].replace("_", " ").title()
-            fields = ", ".join(change.get("changed_fields", []))
-            if fields:
-                detail += f" ({fields})"
-            lines.append(f"| Section | {change['category']} | {change['section']} | - | {detail} |")
-
-        for change in category_changes:
-            detail = change["change_type"].replace("_", " ").title()
-            lines.append(f"| Category | {change['category']} | - | - | {detail} |")
-
-        lines.append("")
+        lines.append("No changes detected in `awesome-privacy.yml`.")
 
     with open(summary_file, "a") as f:
         f.write("\n".join(lines) + "\n")
 
 
 def main():
-    parser = argparse.ArgumentParser(description="Analyze YAML diff for PR checks")
-    parser.add_argument("--base-ref", required=True, help="Base git ref (SHA or branch) to diff against")
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--base-ref", required=True)
     args = parser.parse_args()
 
-    # Load both versions
-    base_data = load_base_yaml(args.base_ref)
-    head_data = load_head_yaml()
+    base = load_base_yaml(args.base_ref)
+    head = load_head_yaml()
 
-    # Compute diffs
-    added, removed, modified = diff_services(base_data, head_data)
-    section_changes = diff_sections(base_data, head_data)
-    category_changes = diff_categories(base_data, head_data)
+    svc_added, svc_removed, svc_modified = diff_index(
+        build_index(base, 3), build_index(head, 3),
+    )
+    sec_added, sec_removed, sec_modified = diff_index(
+        build_index(base, 2), build_index(head, 2),
+    )
+    cat_added, cat_removed, _ = diff_index(
+        build_index(base, 1), build_index(head, 1),
+    )
+
+    added = [{"category": k[0], "section": k[1], "service": k[2],
+              "fields": build_index(head, 3)[k]} for k in svc_added]
+    removed = [{"category": k[0], "section": k[1], "service": k[2]} for k in svc_removed]
+    modified = [{"category": k[0], "section": k[1], "service": k[2],
+                 "changed_fields": cf} for k, cf in svc_modified]
+
+    sections = []
+    for k in sec_added:
+        sections.append({"category": k[0], "section": k[1], "change_type": "added_section"})
+    for k in sec_removed:
+        sections.append({"category": k[0], "section": k[1], "change_type": "removed_section"})
+    for k, cf in sec_modified:
+        sections.append({"category": k[0], "section": k[1],
+                         "change_type": "modified_section_metadata", "changed_fields": cf})
+
+    categories = []
+    for k in cat_added:
+        categories.append({"category": k, "change_type": "added_category"})
+    for k in cat_removed:
+        categories.append({"category": k, "change_type": "removed_category"})
 
-    # Build result
     diff_result = {
-        "services": {
-            "added": added,
-            "removed": removed,
-            "modified": modified,
-        },
-        "sections": section_changes,
-        "categories": category_changes,
+        "services": {"added": added, "removed": removed, "modified": modified},
+        "sections": sections,
+        "categories": categories,
     }
 
-    # Write diff JSON
     with open(DIFF_OUTPUT_PATH, "w") as f:
         json.dump(diff_result, f, indent=2)
-    print(f"Diff written to {DIFF_OUTPUT_PATH}")
 
-    # Determine if there are service-level changes
-    has_service_changes = bool(added or removed or modified)
-    write_github_output("has_service_changes", str(has_service_changes).lower())
-
-    # Write step summary
+    write_github_output("has_service_changes", str(bool(added or removed or modified)).lower())
     write_step_summary(diff_result)
 
-    # Enforce single-entry rule
-    service_change_count = len(added) + len(removed) + len(modified)
-
-    if service_change_count > 1:
-        print(red("Single-entry rule violation: PRs must contain only one service change."), file=sys.stderr)
-        print(red(f"Found {service_change_count} service-level changes:"), file=sys.stderr)
-        for svc in added:
-            print(f"  + Added: {svc['category']} > {svc['section']} > {svc['service']}", file=sys.stderr)
-        for svc in removed:
-            print(f"  - Removed: {svc['category']} > {svc['section']} > {svc['service']}", file=sys.stderr)
-        for svc in modified:
-            fields = ", ".join(svc["changed_fields"])
-            print(f"  ~ Modified: {svc['category']} > {svc['section']} > {svc['service']} ({fields})", file=sys.stderr)
+    svc_count = len(added) + len(removed) + len(modified)
+    if svc_count > 1:
+        print(red(f"Single-entry rule violation: {svc_count} service changes found."), file=sys.stderr)
+        sys.exit(EXIT_RULE_VIOLATION)
+    if svc_count == 0 and len(sections) > 1:
+        print(red(f"Single-entry rule violation: {len(sections)} section changes found."), file=sys.stderr)
         sys.exit(EXIT_RULE_VIOLATION)
 
-    # If no service changes, check section-level changes
-    if service_change_count == 0 and len(section_changes) > 1:
-        print(red("Single-entry rule violation: PRs must contain only one section-level change."), file=sys.stderr)
-        print(red(f"Found {len(section_changes)} section-level changes:"), file=sys.stderr)
-        for change in section_changes:
-            detail = change["change_type"].replace("_", " ")
-            fields = change.get("changed_fields", [])
-            extra = f" ({', '.join(fields)})" if fields else ""
-            print(f"  ~ {change['category']} > {change['section']}: {detail}{extra}", file=sys.stderr)
-        sys.exit(EXIT_RULE_VIOLATION)
-
-    # Summary
-    total = service_change_count + len(section_changes) + len(category_changes)
-    if total == 0:
-        print(green("No changes detected in awesome-privacy.yml"))
-    else:
-        print(green(f"Single-entry rule passed. {service_change_count} service change(s), "
-                     f"{len(section_changes)} section change(s), {len(category_changes)} category change(s)."))
+    print(green(f"Single-entry rule passed. {svc_count} service, "
+                f"{len(sections)} section, {len(categories)} category change(s)."))
     sys.exit(EXIT_PASS)
 
 
diff --git a/lib/checks/detect-changes.py b/lib/checks/detect-changes.py
index 9b87e78..9fa591d 100644
--- a/lib/checks/detect-changes.py
+++ b/lib/checks/detect-changes.py
@@ -1,6 +1,6 @@
 """
 Detects which files changed between the PR base and HEAD.
-Sets GitHub Actions outputs: yaml_changed, non_yaml_changed.
+Sets GitHub Actions output: yaml_changed.
 """
 
 import argparse
@@ -38,12 +38,8 @@ def main():
         print(f"  {f}")
 
     yaml_changed = YAML_FILE in changed_files
-    non_yaml_changed = any(f != YAML_FILE for f in changed_files)
-
     write_github_output("yaml_changed", str(yaml_changed).lower())
-    write_github_output("non_yaml_changed", str(non_yaml_changed).lower())
-
-    print(f"yaml_changed={yaml_changed}, non_yaml_changed={non_yaml_changed}")
+    print(f"yaml_changed={yaml_changed}")
 
 
 if __name__ == "__main__":
diff --git a/lib/checks/format-comment.py b/lib/checks/format-comment.py
index e81497a..d2789b1 100644
--- a/lib/checks/format-comment.py
+++ b/lib/checks/format-comment.py
@@ -7,12 +7,6 @@ import sys
 ARTIFACTS_DIR = "/tmp/artifacts"
 OUTPUT_DIR = "/tmp/pr-meta"
 
-README_MSG = (
-    "Do not edit the README directly. This file is auto-generated from the"
-    " content in `awesome-privacy.yml`, and so your changes will be overridden!"
-    " Instead, only modify the YAML file, and be sure to follow our Contributing Guidelines."
-)
-
 CONTRIBUTING = "https://github.com/Lissy93/awesome-privacy/blob/main/.github/CONTRIBUTING.md"
 
 COMMENT_TEMPLATE = """<!-- pr-check-bot -->
@@ -38,12 +32,10 @@ def load_findings(filename):
         return []
 
 
-def collect_findings(readme_failed):
-    """Gather all findings in display order: meta, readme, data, project."""
+def collect_findings():
+    """Gather all findings in display order: compliance, data, project."""
     all_findings = []
-    all_findings.extend(load_findings("findings-meta.json"))
-    if readme_failed:
-        all_findings.append(README_MSG)
+    all_findings.extend(load_findings("findings-compliance.json"))
     all_findings.extend(load_findings("findings-data.json"))
     all_findings.extend(load_findings("findings-project.json"))
     return all_findings
@@ -78,7 +70,6 @@ def main():
         user = os.environ.get("PR_USER", "contributor")
         pr_number = os.environ.get("PR_NUMBER", "")
         run_id = os.environ.get("RUN_ID", "")
-        readme_failed = os.environ.get("README_FAILED", "false") == "true"
 
         os.makedirs(OUTPUT_DIR, exist_ok=True)
 
@@ -89,7 +80,7 @@ def main():
             with open(os.path.join(OUTPUT_DIR, "run-id.txt"), "w") as f:
                 f.write(run_id)
 
-        findings = collect_findings(readme_failed)
+        findings = collect_findings()
         write_step_summary(findings)
 
         if findings:
diff --git a/lib/checks/warn-non-yaml.py b/lib/checks/warn-non-yaml.py
deleted file mode 100644
index e244dbc..0000000
--- a/lib/checks/warn-non-yaml.py
+++ /dev/null
@@ -1,58 +0,0 @@
-"""
-Warns when a PR modifies files other than awesome-privacy.yml.
-This is expected for Website Update or Misc PRs, but may need extra review.
-"""
-
-import argparse
-import os
-import subprocess
-import sys
-
-PROJECT_ROOT = os.path.dirname(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
-
-YAML_FILE = "awesome-privacy.yml"
-
-# ANSI color helpers
-_use_color = sys.stderr.isatty() and not os.environ.get("NO_COLOR")
-yellow = (lambda s: f"\033[33m{s}\033[0m") if _use_color else (lambda s: s)
-
-
-def main():
-    parser = argparse.ArgumentParser(description="Warn about non-YAML file changes")
-    parser.add_argument("--base-ref", required=True, help="Base git ref to diff against")
-    args = parser.parse_args()
-
-    result = subprocess.run(
-        ["git", "diff", "--name-only", f"{args.base_ref}..HEAD"],
-        capture_output=True, text=True, check=True,
-        cwd=PROJECT_ROOT,
-    )
-
-    changed_files = [f for f in result.stdout.strip().splitlines() if f]
-    non_yaml = [f for f in changed_files if f != YAML_FILE]
-
-    if not non_yaml:
-        return
-
-    print(yellow("This PR modifies files other than awesome-privacy.yml:"), file=sys.stderr)
-    for f in non_yaml:
-        print(f"  {f}", file=sys.stderr)
-
-    # Write step summary
-    summary_file = os.environ.get("GITHUB_STEP_SUMMARY")
-    if summary_file:
-        lines = [
-            "## Non-YAML Changes Warning\n",
-            "This PR modifies files other than `awesome-privacy.yml`:\n",
-        ]
-        for f in non_yaml:
-            lines.append(f"- `{f}`")
-        lines.append("")
-        lines.append("> **Note:** Most PRs should only modify `awesome-privacy.yml`. "
-                      "Non-YAML changes may require additional review.\n")
-        with open(summary_file, "a") as f:
-            f.write("\n".join(lines) + "\n")
-
-
-if __name__ == "__main__":
-    main()