weval-l99/agent-scanner.py

#!/usr/bin/env python3
"""WEVIA AGENT SCANNER v1.0 — Deep Architecture Scanner + Wiki Updater
Scans: servers, Docker, APIs, routes, skills, crons, Qdrant, providers, pages, OSS
Updates wiki automatically. Cron */2h."""
import json, os, subprocess, urllib.request, ssl, glob
from datetime import datetime

ssl._create_default_https_context = ssl._create_unverified_context
WIKI_DIR = "/opt/weval-l99/wiki"
REPORT = "/var/www/html/api/agent-scanner-report.json"
MM_HOOK = "http://localhost:8065/hooks/pt54hzthf3b6pe6rgp1ionipnh"
os.makedirs(WIKI_DIR, exist_ok=True)

def log(m): print(f"[{datetime.now().strftime('%H:%M:%S')}] {m}")
def sh(c): return subprocess.run(c, shell=True, capture_output=True, text=True, timeout=10).stdout.strip()
def wiki(key, data):
    with open(f"{WIKI_DIR}/{key}.json", "w") as f: json.dump({"key": key, "updated": datetime.now().isoformat(), "data": data}, f, ensure_ascii=False, indent=2)

def main():
    log("=== AGENT SCANNER v1.0 ===")
    report = {"timestamp": datetime.now().strftime("%Y-%m-%d %H:%M"), "sections": {}}

    # 1. SERVERS
    log("1. Servers")
    s204 = {"docker": int(sh("docker ps | tail -n+2 | wc -l") or 0), "disk": sh("df -h / | tail -1 | awk '{print $5}'"),
            "ram": sh("free -h | grep Mem | awk '{print $3\"/\"$2}'"), "load": sh("cat /proc/loadavg | head -c 5"),
            "uptime": sh("uptime -p")}
    report["sections"]["servers"] = {"S204": s204}
    wiki("SERVERS", s204)

    # 2. DOCKER CONTAINERS
    log("2. Docker")
    containers = []
    for line in sh("docker ps --format '{{.Names}}|{{.Status}}|{{.Ports}}'").split("\n"):
        if line:
            parts = line.split("|")
            containers.append({"name": parts[0], "status": parts[1] if len(parts) > 1 else "?", "ports": parts[2] if len(parts) > 2 else ""})
    report["sections"]["docker"] = {"count": len(containers), "containers": containers}
    wiki("DOCKER", {"count": len(containers), "names": [c["name"] for c in containers]})

    # 3. APIS
    log("3. APIs")
    apis = [os.path.basename(f) for f in glob.glob("/var/www/html/api/wevia-*.php") + glob.glob("/var/www/html/api/weval-*.php")]
    report["sections"]["apis"] = {"count": len(apis), "files": apis}
    wiki("APIS", {"count": len(apis), "list": apis})

    # 4. ROUTES
    log("4. Routes")
    fast = "/var/www/html/api/weval-ia-fast.php"
    if os.path.exists(fast):
        lines = sum(1 for _ in open(fast))
        routes = sum(1 for l in open(fast) if "// Route" in l)
        report["sections"]["routes"] = {"lines": lines, "count": routes}
        wiki("ROUTES", {"lines": lines, "count": routes})

    # 5. SKILLS
    log("5. Skills")
    skills = [os.path.basename(d) for d in glob.glob("/opt/deer-flow/skills/weval/*/")]
    report["sections"]["skills"] = {"count": len(skills)}
    wiki("SKILLS", {"count": len(skills), "latest": skills[-5:] if skills else []})

    # 6. CRONS
    log("6. Crons")
    crons = glob.glob("/etc/cron.d/weval-*")
    report["sections"]["crons"] = {"count": len(crons), "files": [os.path.basename(c) for c in crons]}
    wiki("CRONS", {"count": len(crons), "list": [os.path.basename(c) for c in crons]})

    # 7. QDRANT
    log("7. Qdrant")
    try:
        r = urllib.request.urlopen("http://127.0.0.1:6333/collections", timeout=5).read()
        d = json.loads(r)
        colls = {}
        for c in d.get("result", {}).get("collections", []):
            name = c.get("name", "?")
            try:
                r2 = urllib.request.urlopen(f"http://127.0.0.1:6333/collections/{name}", timeout=3).read()
                d2 = json.loads(r2)
                colls[name] = d2.get("result", {}).get("points_count", 0)
            except: colls[name] = "?"
        total = sum(v for v in colls.values() if isinstance(v, int))
        report["sections"]["qdrant"] = {"total": total, "collections": colls}
        wiki("QDRANT", {"total": total, "collections": colls})
    except: report["sections"]["qdrant"] = "unavailable"

    # 8. OLLAMA
    log("8. Ollama")
    try:
        r = urllib.request.urlopen("http://127.0.0.1:11434/api/tags", timeout=5).read()
        d = json.loads(r)
        models = [m.get("name", "?") for m in d.get("models", [])]
        report["sections"]["ollama"] = {"count": len(models), "models": models}
        wiki("OLLAMA", {"count": len(models), "models": models})
    except: report["sections"]["ollama"] = "unavailable"

    # 9. HTML PAGES
    log("9. Pages")
    pages = [os.path.basename(f) for f in glob.glob("/var/www/html/*.html")]
    report["sections"]["pages"] = {"count": len(pages)}
    wiki("PAGES", {"count": len(pages)})

    # 10. OPT TOOLS
    log("10. /opt tools")
    tools = [os.path.basename(d) for d in glob.glob("/opt/*/") if os.path.isdir(d)]
    report["sections"]["opt_tools"] = {"count": len(tools)}
    wiki("OPT_TOOLS", {"count": len(tools), "list": tools})

    # 11. DATASET
    log("11. Dataset")
    ds_file = "/opt/wevia-brain/training-data/weval-merged-dataset.jsonl"
    ds_count = int(sh(f"wc -l {ds_file} | awk '{{print $1}}'") or 0) if os.path.exists(ds_file) else 0
    report["sections"]["dataset"] = {"pairs": ds_count}
    wiki("DATASET", {"pairs": ds_count})

    # 12. WIKI STATS
    log("12. Wiki")
    wiki_files = glob.glob(f"{WIKI_DIR}/*.json")
    report["sections"]["wiki"] = {"entries": len(wiki_files)}

    # SAVE
    with open(REPORT, "w") as f: json.dump(report, f, indent=2, ensure_ascii=False)
    log(f"Report saved: {REPORT}")
    log(f"Wiki updated: {len(wiki_files)} entries")

    # Summary
    sections = report["sections"]
    log(f"DONE: Docker={sections.get('docker',{}).get('count','?')} Routes={sections.get('routes',{}).get('count','?')} Skills={sections.get('skills',{}).get('count','?')} Crons={sections.get('crons',{}).get('count','?')} Wiki={sections.get('wiki',{}).get('entries','?')}")

try: main()
except Exception as e: print(f"CRASH: {e}"); import traceback; traceback.print_exc()