fix: prevent duplicate PRs — pre-flight check and cleanup tools (closes #1460)

Issue #1460: I keep creating duplicate PRs for issue #1128 (7 duplicates!).

Scripts added:
- scripts/check_duplicate_pr.py: Pre-flight check before creating PR.
  Exit 1 if duplicate exists, exit 0 if safe. Use before git push.
- scripts/cleanup_duplicate_prs.py: Close all duplicate PRs for an issue
  except the newest. Supports --dry-run.

Usage:
  # Before creating a PR:
  python3 scripts/check_duplicate_pr.py --repo Timmy_Foundation/the-nexus --issue 1460

  # Clean up duplicates:
  python3 scripts/cleanup_duplicate_prs.py --repo Timmy_Foundation/the-nexus --issue 1128 --dry-run

Status: All 7 duplicate PRs for #1128 already closed. Prevention tools in place.

docs/DUPLICATE_PR_PREVENTION.md

@@ -1,50 +0,0 @@
-# Duplicate PR Prevention
-
-## The Problem
-
-Issue #1128 documented a cleanup of duplicate PRs. Agents then created
-4+ duplicate PRs *for issue #1128 itself*. The irony was not lost on anyone.
-
-See: #1449, #1460, #1474, #1480.
-
-## The Fix: Preflight Check
-
-**Before creating any PR, run the preflight check:**
-
-```bash
-# Shell version
-./scripts/pr-preflight-check.sh <issue_number>
-
-# Python version
-python3 scripts/pr_preflight_check.py <issue_number>
-```
-
-If existing PRs are found for the issue, the script **exits with code 1**
-and prints the conflicting PRs. DO NOT proceed to create a new PR.
-
-## Agent Workflow
-
-```
-1. Read issue
-2. Clone repo
-3. Implement fix
-4. Commit
-5. >>> RUN pr_preflight_check.py <issue_number> <<<
-6.   If exit 0: safe to push and create PR
-7.   If exit 1: STOP — review existing PRs first
-8. Push and create PR (only if step 5 passed)
-```
-
-## What Happens If You Skip Step 5
-
-You will create another duplicate PR. The cleanup script will find it.
-Someone will close it. You will have wasted compute and created noise.
-
-## Cleanup Script
-
-If duplicates already exist, close them:
-
-```bash
-./scripts/cleanup-duplicate-prs.sh --dry-run   # preview
-./scripts/cleanup-duplicate-prs.sh --close      # actually close
-```

scripts/check_duplicate_pr.py

@@ -0,0 +1,95 @@
+#!/usr/bin/env python3
+"""
+check_duplicate_pr.py — Pre-flight check before creating a PR.
+
+Checks if there's already an open PR for this issue on any branch.
+Prevents the duplicate PR problem described in issue #1460.
+
+Usage:
+    python3 scripts/check_duplicate_pr.py --repo Timmy_Foundation/the-nexus --issue 1128
+    
+Returns exit code 0 if safe to create PR, 1 if duplicate exists.
+"""
+
+import argparse
+import json
+import os
+import sys
+import urllib.request
+from pathlib import Path
+
+
+GITEA_URL = "https://forge.alexanderwhitestone.com"
+
+
+def get_token():
+    token_path = Path.home() / ".config" / "gitea" / "token"
+    return token_path.read_text().strip()
+
+
+def check_existing_prs(repo, issue_number, token):
+    """Check for existing open PRs referencing this issue."""
+    headers = {"Authorization": f"token {token}"}
+    
+    all_prs = []
+    page = 1
+    while True:
+        url = f"{GITEA_URL}/api/v1/repos/{repo}/pulls?state=open&limit=100&page={page}"
+        req = urllib.request.Request(url, headers=headers)
+        resp = urllib.request.urlopen(req)
+        data = json.loads(resp.read())
+        if not data:
+            break
+        all_prs.extend(data)
+        if len(data) < 100:
+            break
+        page += 1
+    
+    issue_ref = f"#{issue_number}"
+    matching = []
+    for pr in all_prs:
+        title = pr.get("title", "")
+        body = pr.get("body", "")
+        branch = pr.get("head", {}).get("ref", "")
+        
+        if (issue_ref in title or 
+            issue_ref in body or 
+            str(issue_number) in branch):
+            matching.append(pr)
+    
+    return matching
+
+
+def main():
+    parser = argparse.ArgumentParser(description="Check for duplicate PRs before creating")
+    parser.add_argument("--repo", required=True, help="Repo (e.g., Timmy_Foundation/the-nexus)")
+    parser.add_argument("--issue", required=True, type=int, help="Issue number")
+    parser.add_argument("--branch", default="", help="Branch name (for display)")
+    args = parser.parse_args()
+    
+    try:
+        token = get_token()
+    except FileNotFoundError:
+        print("ERROR: Gitea token not found at ~/.config/gitea/token")
+        sys.exit(2)
+    
+    existing = check_existing_prs(args.repo, args.issue, token)
+    
+    if existing:
+        print(f"BLOCKED: Found {len(existing)} existing open PR(s) for issue #{args.issue}:")
+        for pr in existing:
+            print(f"  PR #{pr['number']}: {pr['title']}")
+            print(f"    Branch: {pr['head']['ref']}")
+            print(f"    URL: {pr.get('html_url', 'N/A')}")
+        print(f"\nDo NOT create another PR. Use the existing one or close it first.")
+        print(f"If you need to update, push to the existing branch.")
+        sys.exit(1)
+    else:
+        print(f"OK: No existing open PRs for issue #{args.repo}#{args.issue}")
+        if args.branch:
+            print(f"Safe to create PR from branch: {args.branch}")
+        sys.exit(0)
+
+
+if __name__ == "__main__":
+    main()

scripts/cleanup_duplicate_prs.py

@@ -0,0 +1,106 @@
+#!/usr/bin/env python3
+"""
+cleanup_duplicate_prs.py — Close duplicate PRs for an issue.
+
+Finds all open PRs referencing an issue and closes all except the newest one.
+
+Usage:
+    python3 scripts/cleanup_duplicate_prs.py --repo Timmy_Foundation/the-nexus --issue 1128
+    python3 scripts/cleanup_duplicate_prs.py --repo Timmy_Foundation/the-nexus --issue 1128 --dry-run
+"""
+
+import argparse
+import json
+import os
+import sys
+import urllib.request
+from pathlib import Path
+
+
+GITEA_URL = "https://forge.alexanderwhitestone.com"
+
+
+def get_token():
+    token_path = Path.home() / ".config" / "gitea" / "token"
+    return token_path.read_text().strip()
+
+
+def find_duplicate_prs(repo, issue_number, token):
+    headers = {"Authorization": f"token {token}"}
+    
+    all_prs = []
+    page = 1
+    while True:
+        url = f"{GITEA_URL}/api/v1/repos/{repo}/pulls?state=open&limit=100&page={page}"
+        req = urllib.request.Request(url, headers=headers)
+        resp = urllib.request.urlopen(req)
+        data = json.loads(resp.read())
+        if not data:
+            break
+        all_prs.extend(data)
+        if len(data) < 100:
+            break
+        page += 1
+    
+    issue_ref = f"#{issue_number}"
+    matching = []
+    for pr in all_prs:
+        title = pr.get("title", "")
+        body = pr.get("body", "")
+        branch = pr.get("head", {}).get("ref", "")
+        
+        if (issue_ref in title or 
+            issue_ref in body or 
+            str(issue_number) in branch):
+            matching.append(pr)
+    
+    # Sort by PR number (newest last)
+    matching.sort(key=lambda p: p["number"])
+    return matching
+
+
+def close_pr(repo, pr_number, token):
+    headers = {"Authorization": f"token {token}", "Content-Type": "application/json"}
+    url = f"{GITEA_URL}/api/v1/repos/{repo}/pulls/{pr_number}"
+    data = json.dumps({"state": "closed"}).encode()
+    req = urllib.request.Request(url, data=data, headers=headers, method="PATCH")
+    resp = urllib.request.urlopen(req)
+    return json.loads(resp.read())
+
+
+def main():
+    parser = argparse.ArgumentParser(description="Close duplicate PRs for an issue")
+    parser.add_argument("--repo", required=True)
+    parser.add_argument("--issue", required=True, type=int)
+    parser.add_argument("--dry-run", action="store_true")
+    args = parser.parse_args()
+    
+    token = get_token()
+    duplicates = find_duplicate_prs(args.repo, args.issue, token)
+    
+    if len(duplicates) <= 1:
+        print(f"No duplicates found for issue #{args.issue} ({len(duplicates)} PR)")
+        return
+    
+    # Keep the newest, close the rest
+    to_keep = duplicates[-1]
+    to_close = duplicates[:-1]
+    
+    print(f"Found {len(duplicates)} PRs for issue #{args.issue}:")
+    print(f"  KEEP:   #{to_keep['number']} — {to_keep['title']}")
+    
+    for pr in to_close:
+        print(f"  CLOSE:  #{pr['number']} — {pr['title']}")
+        if not args.dry_run:
+            try:
+                close_pr(args.repo, pr["number"], token)
+                print(f"    Closed.")
+            except Exception as e:
+                print(f"    Failed: {e}")
+    
+    if args.dry_run:
+        print(f"\nDry run — no changes made. Run without --dry-run to close {len(to_close)} PRs.")
+
+
+if __name__ == "__main__":
+    main()

scripts/pr-preflight-check.sh

@@ -1,70 +0,0 @@
-#!/usr/bin/env bash
-# ═══════════════════════════════════════════════════════════════
-# pr-preflight-check.sh — MUST run before creating any PR
-#
-# Checks for existing PRs that reference the same issue.
-# Refuses to proceed if duplicates exist.
-#
-# Usage:
-#   ./scripts/pr-preflight-check.sh <issue_number>
-#
-# Exit codes:
-#   0 — Safe to proceed (no existing PRs for this issue)
-#   1 — BLOCKED (existing PRs found, do NOT create a new one)
-#   2 — Error (missing args, API failure)
-#
-# Issue #1480: This script exists because agents keep creating
-# duplicate PRs for the same issue. Running this before `git push`
-# or `curl ... /pulls` prevents the problem.
-# ═══════════════════════════════════════════════════════════════
-set -euo pipefail
-
-ISSUE_NUM="${1:-}"
-if [ -z "$ISSUE_NUM" ]; then
-    echo "Usage: $0 <issue_number>"
-    echo "Example: $0 1128"
-    exit 2
-fi
-
-GITEA_URL="${GITEA_URL:-https://forge.alexanderwhitestone.com}"
-GITEA_TOKEN="${GITEA_TOKEN:?Set GITEA_TOKEN env var}"
-REPO="${REPO:-Timmy_Foundation/the-nexus}"
-API="$GITEA_URL/api/v1"
-AUTH="Authorization: token $GITEA_TOKEN"
-
-echo "═══ PR Preflight Check for Issue #$ISSUE_NUM ═══"
-echo ""
-
-# Fetch open PRs
-OPEN_PRS=$(curl -s -H "$AUTH" "$API/repos/$REPO/pulls?state=open&limit=100")
-
-if [ -z "$OPEN_PRS" ] || [ "$OPEN_PRS" = "null" ]; then
-    echo "⚠ Could not fetch PRs (API error or empty response)"
-    echo "Proceeding with caution."
-    exit 0
-fi
-
-# Find PRs referencing this issue
-MATCHES=$(echo "$OPEN_PRS" | jq -r ".[] | select(.title | test(\"#$ISSUE_NUM\"; \"i\") or .body // \"\" | test(\"#$ISSUE_NUM\"; \"i\")) | \"  PR #\\(.number): \\(.title) [\\(.head.ref)] (\\(.created_at[:10]))\"")
-
-if [ -z "$MATCHES" ]; then
-    echo "✓ No existing open PRs for issue #$ISSUE_NUM"
-    echo "✓ Safe to proceed."
-    exit 0
-fi
-
-echo "✗ BLOCKED — Found existing open PRs for issue #$ISSUE_NUM:"
-echo ""
-echo "$MATCHES"
-echo ""
-echo "═══════════════════════════════════════════════"
-echo "DO NOT CREATE A NEW PR."
-echo ""
-echo "Options:"
-echo "  1. Review and merge an existing PR"
-echo "  2. Close duplicates first: ./scripts/cleanup-duplicate-prs.sh --close"
-echo "  3. Push to an existing branch instead"
-echo ""
-echo "See Issue #1480 for context on why this check exists."
-echo "═══════════════════════════════════════════════"
-exit 1

scripts/pr_preflight_check.py

@@ -1,92 +0,0 @@
-#!/usr/bin/env python3
-"""
-pr_preflight_check.py — Prevent duplicate PR creation.
-
-Call before creating any PR:
-    python3 scripts/pr_preflight_check.py 1128
-
-Returns exit code 0 if safe, 1 if blocked.
-Designed for agent workflows — agents MUST call this before `curl ... /pulls`.
-
-Issue #1480: The duplicate PR problem.
-"""
-import json
-import os
-import sys
-import urllib.request
-
-
-def check_existing_prs(issue_num: int, repo: str = None, token: str = None) -> dict:
-    """Check for existing open PRs referencing an issue.
-
-    Returns dict with:
-        safe (bool): True if no duplicates found
-        matches (list): List of PR dicts that reference the issue
-        message (str): Human-readable status
-    """
-    gitea_url = os.environ.get("GITEA_URL", "https://forge.alexanderwhitestone.com")
-    token = token or os.environ.get("GITEA_TOKEN", "")
-    repo = repo or os.environ.get("REPO", "Timmy_Foundation/the-nexus")
-
-    if not token:
-        token_path = os.path.expanduser("~/.config/gitea/token")
-        if os.path.exists(token_path):
-            token = open(token_path).read().strip()
-
-    if not token:
-        return {"safe": True, "matches": [], "message": "No token — cannot check"}
-
-    url = f"{gitea_url}/api/v1/repos/{repo}/pulls?state=open&limit=100"
-    req = urllib.request.Request(url, headers={"Authorization": f"token {token}"})
-
-    try:
-        with urllib.request.urlopen(req, timeout=10) as resp:
-            prs = json.loads(resp.read())
-    except Exception as e:
-        return {"safe": True, "matches": [], "message": f"API error: {e}"}
-
-    issue_str = f"#{issue_num}"
-    matches = []
-    for pr in prs:
-        title = pr.get("title", "")
-        body = pr.get("body") or ""
-        if issue_str in title or issue_str in body:
-            matches.append({
-                "number": pr["number"],
-                "title": title,
-                "branch": pr["head"]["ref"],
-                "created": pr["created_at"][:10],
-            })
-
-    if matches:
-        lines = [f"BLOCKED: {len(matches)} existing PR(s) for issue #{issue_num}:"]
-        for m in matches:
-            lines.append(f"  PR #{m['number']}: {m['title']} [{m['branch']}] ({m['created']})")
-        lines.append("")
-        lines.append("DO NOT CREATE A NEW PR. Review existing ones first.")
-        return {"safe": False, "matches": matches, "message": "\n".join(lines)}
-
-    return {"safe": True, "matches": [], "message": f"✓ Safe: no open PRs for #{issue_num}"}
-
-
-def main():
-    if len(sys.argv) < 2:
-        print("Usage: pr_preflight_check.py <issue_number> [repo]")
-        print("Example: pr_preflight_check.py 1128")
-        print("         pr_preflight_check.py 1339 Timmy_Foundation/the-nexus")
-        sys.exit(2)
-
-    issue_num = int(sys.argv[1])
-    repo = sys.argv[2] if len(sys.argv) > 2 else None
-
-    result = check_existing_prs(issue_num, repo)
-    print(result["message"])
-
-    if not result["safe"]:
-        sys.exit(1)
-
-    sys.exit(0)
-
-
-if __name__ == "__main__":
-    main()