live · scanning repos
Scanrepo
github.com
DANGER
malware detected
github.com / voidmatcha

voidmatcha/ui-clone-skills

Python·914 files·commit c6dacc7·scanned 12h ago·cached ✓
85/100
MALICIOUS
Matches known-malicious patterns. Do not clone or install.

This repository contains patterns associated with malware. Do NOT run this code.

verdict accurate?
Partial architecture graph
CLI tool detected
threat-state: maliciouslive

FINDINGS ░▒▓

what this code does░▒▓

What happens when you open, install or run it, from the findings below.

  1. when runscripts/hooks/post-push-refresh.sh downloads a script and pipes it to the shell. scripts/hooks/post-push-refresh.sh
  2. when runscripts/analysis/context_token_attribution.py and scripts/extract/_inline_scripts.py build code from strings at run time with exec or eval. scripts/analysis/context_token_attribution.py, scripts/extract/_inline_scripts.py
  3. when runskills/visual-debug/scripts/visual_fidelity_judge.py starts system processes. skills/visual-debug/scripts/visual_fidelity_judge.py
criticaleval() usage detectedeval() executes arbitrary code and is commonly used in malware to run obfuscated payloads fetched from remote servers.scripts/analysis/context_token_attribution.py
_AGENT_BROWSER_RE = re.compile(r"(?<![\w.-])(?:[^\s;&|]*/)?agent-browser(?:\s|$)")
_EVAL_RE = re.compile(r"(?<![\w.-])eval(?:\s|$)")
_DIRECT_READ_RE = re.compile(r"(?<![\w.-])(?:cat|sed|head|tail|nl|j...
+8
criticalPython dynamic code executioneval(), exec() or compile() on untrusted input can execute arbitrary code. Common in Python malware for running obfuscated payloads.scripts/analysis/context_token_attribution.py
_AGENT_BROWSER_RE = re.compile(r"(?<![\w.-])(?:[^\s;&|]*/)?agent-browser(?:\s|$)")
_EVAL_RE = re.compile(r"(?<![\w.-])eval(?:\s|$)")
_DIRECT_READ_RE = re.compile(r"(?<![\w.-])(?:cat|sed|head|tail|nl|j...
+7
criticaleval() usage detectedeval() executes arbitrary code and is commonly used in malware to run obfuscated payloads fetched from remote servers.scripts/ci/pre-push-security.sh
  grep -vE "^[^:]*:[0-9]+:[[:space:]]*#" | wc -l | tr -d ' ')
[ "$eval_count" -eq 0 ] && ok "no bash eval()" || err "bash eval() found ($eval_count occurrences)"

# CWE-377: insecure use of fixed temp...
+8
criticaleval() usage detectedeval() executes arbitrary code and is commonly used in malware to run obfuscated payloads fetched from remote servers.scripts/extract/_inline_scripts.py
    except json.JSONDecodeError as exc:
        print(f"inline-scripts: invalid JSON from agent-browser eval ({exc})", file=sys.stderr)
        return 3
+8
criticalPython dynamic code executioneval(), exec() or compile() on untrusted input can execute arbitrary code. Common in Python malware for running obfuscated payloads.scripts/extract/_inline_scripts.py
    except json.JSONDecodeError as exc:
        print(f"inline-scripts: invalid JSON from agent-browser eval ({exc})", file=sys.stderr)
        return 3
+7
criticalRemote script piped into a shellA shell script downloads remote content and executes it directly (e.g. `curl … | bash`). The payload is never stored or reviewable and can change server-side — a classic malware/backdoor install pattern.scripts/hooks/post-push-refresh.sh
# Agent post-push refresh hook: after a successful `git push`, wipe the local
# install at INSTALL_DIR and re-run the canonical curl-pipe install. This
# dogfoods install.sh end-to-end against the jus...
+10
warningPython subprocess / os.system usageos.system() or subprocess with shell=True / suspicious commands can execute arbitrary shell commands. Plain pip/package installs are usually benign.skills/visual-debug/scripts/visual_fidelity_judge.py
        try:
            proc = subprocess.run(["bash", str(resolver), str(REF_DIR)],
                                  capture_output=True, text=True, timeout=30)
        except Exception:
+4
criticalPython dynamic code executioneval(), exec() or compile() on untrusted input can execute arbitrary code. Common in Python malware for running obfuscated payloads.tests/gates/test_content_cardinality.py
globalThis.document = { body: parent, querySelectorAll: () => [parent] };
const out = eval(fs.readFileSync(process.argv[2], 'utf8'));
console.log(out);
""",
+7
warningPython subprocess / os.system usageos.system() or subprocess with shell=True / suspicious commands can execute arbitrary shell commands. Plain pip/package installs are usually benign.tests/gates/test_hover_affected_selector_absent.py
    hover_css.write_text(json.dumps(HOVER_CSS), encoding="utf-8")
    proc = subprocess.run(
        ["bash", "-c", _function_source() + '\naffected_selector_for_hover "$1"', "_", activation],
       ...
+4
warningPython subprocess / os.system usageos.system() or subprocess with shell=True / suspicious commands can execute arbitrary shell commands. Plain pip/package installs are usually benign.tests/gates/test_hover_state_cap.py
    try:
        common = subprocess.run(
            ["git", "-C", str(ROOT), "rev-parse", "--git-common-dir"],
            capture_output=True, text=True, check=True, timeout=30,
+4
criticalFunction() constructor detectednew Function() is equivalent to eval() and can execute arbitrary code strings.tests/measure/opener_walk_harness.js
// user input); evaluating it is the point of the harness.
const api = new Function(lib + '\nreturn { tfRefusal, tfPickOpener, tfOpenerWalk, tfOpenerClose, tfFingerprint, tfSettled };')();

const togg...
+5
infoSuspicious files are not reachable from entry pointsFlagged files exist but are not imported by any entry point. They may be dead code, tests, or attack payloads triggered by another mechanism.+2
infoBase64 encoding usedatob() or Buffer.from with base64. Common in legitimate code but also used to hide malicious URLs.skills/visual-debug/scripts/live-parity-sweep.sh
      return src.slice(0, src.indexOf(",")).includes(";base64")
        ? atob(payload)
        : decodeURIComponent(payload);
    }
+2
infoBase64 encoding usedatob() or Buffer.from with base64. Common in legitimate code but also used to hide malicious URLs.skills/visual-debug/scripts/reveal-trigger-check.sh
PROBE_COUNT_RAW=$(agent-browser --session "$SESSION" eval '(() => {
  const SPEC_SELECTORS = JSON.parse(atob("'"$SPEC_SELECTORS_B64"'"));
  const hasScopedLegacy = Array.isArray(SPEC_SELECTORS) && SPE...
+2
infoBase64 encoding usedatob() or Buffer.from with base64. Common in legitimate code but also used to hide malicious URLs.skills/visual-debug/scripts/transition-fires-check.sh
PHASE1="(async () => {
  const ENTRIES = JSON.parse(atob('$ENTRIES_B64'));
  $SNAP_JS
  // Mount sweep (Fix 78c) — React impls conditionally MOUNT animated content
+2
warningExtremely long lines (>1000 chars)Very long lines in source files (not minified bundles) can hide malicious code.ui_clone/section_capture_js.py+4
warningHigh-entropy string literalsFound 6 long strings with high Shannon entropy. This is common in obfuscated payloads that hide URLs, keys, or bytecode.ui_clone/extraction_artifacts.py+3
warningHigh-entropy string literalsFound 3 long strings with high Shannon entropy. This is common in obfuscated payloads that hide URLs, keys, or bytecode.ui_clone/gates/post_implement.py+3
warningHigh-entropy string literalsFound 30 long strings with high Shannon entropy. This is common in obfuscated payloads that hide URLs, keys, or bytecode.ui_clone/hooks/pre_bash_rules/bash_write.py+3

Scores are heuristics. A “safe” verdict means no known-malicious patterns were found — clever malware can look boring. Wrong verdict? Flag it above; confirmed false positives become regression tests.

754 files scanned @ c6dacc7 | 10/30/2026 | heuristic scan — always review manually

risk by category
code execution25
network & exfiltration8
file system access0
obfuscation15
supply chain0
owasp / injection0
telemetry
files 754/914rules hit 19engine v10commit c6dacc7

github

voidmatcha/ui-clone-skills

Motion forensics for the animated web. A Claude Code & Codex plugin that reverse-engineers a live URL into React + Tailwind — pulling real CSS + animation params from JS bundles (GSAP/Framer Motion/Webflow IX2) and verifying with AE/SSIM diff, not screenshots.

Python
16
2
232d
914 files
754 scanned(82%)
c6dacc7

network indicators░▒▓

Hosts this code names or would connect to. Top 50 of 103, decoded and call-site destinations first. Informational, not scored.

architecture░▒▓

entry (0) flagged (174) pkg (151)
551 nodes · 1627 edgesscroll to zoom · click node to jump to finding