Security audit
Security checks for vulnerabilities and agentic risk
This skill is a disclosed Linx file-upload helper that matches its stated purpose and does not show hidden or unrelated behavior.
Install this only if you want an agent to upload specific local files to your Linx instance. Use HTTPS for non-local servers, keep the API key and deletion-key logfile private, and do not ask it to upload credentials or sensitive files unless you intend to share them.
Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.
__pycache__/
*.py[cod]
.env
.env.*
*.conf
uploads.json
Model output is used without validation or sanitization. Unvalidated output injected into downstream contexts (SQL, shell, HTML) enables injection attacks and arbitrary code execution.
headers.chmod(0o600)
target = config["siteurl"] + "upload/" + quote(args.name or file.name, safe="")
response = Path(directory) / "response.json"
result = subprocess.run([
curl, "--disable", "--silent", "--show-error", "--globoff",
"--proto", "=http,https", "--connect-timeout", "30", "--max-time", "300",
"--header", "@" + str(headers), "--upload-file", str(file),
Code enumerates, copies, or searches environment variables for secrets. Bulk environment access can collect credentials unrelated to the skill's stated purpose.
thread.start()
self.addCleanup(server.server_close)
self.addCleanup(server.shutdown)
self.env = {k: v for k, v in os.environ.items() if not k.startswith("LINX_")}
self.env.update(PATH=str(self.bin), HOME=str(self.root), XDG_CONFIG_HOME=str(self.root / "config"),
XDG_STATE_HOME=str(self.root / "state"), LINX_URL=self.base, LINX_API_KEY="environment-key")
self.log = self.root / "state/openclaw-linx/uploads.json"
Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.
export LINX_CONFIG="$HOME/.config/linx-client.conf"
chmod 600 "$LINX_CONFIG"
Example config (replace placeholders locally; never commit real settings):
The skill invokes Python, shell commands, local file access, environment-based secrets, and outbound network uploads, but it does not declare any explicit tool scope or permission boundaries. This creates a real least-privilege and governance gap: an agent may use broad capabilities without transparent constraints, increasing the chance of unintended file disclosure or misuse of configured credentials during uploads.
subprocess module calls execute external commands. Without careful input validation, this enables command injection.
command += ["-f", args.name]
if args.direct:
command += ["-selif"]
result = subprocess.run(command + [str(file)], stdin=subprocess.DEVNULL, capture_output=True, text=True)
lines = result.stdout.strip().splitlines()
# The client can report an HTTP error on stdout while exiting zero.
if result.returncode or len(lines) != 1 or not valid_url(lines[0]):
subprocess module calls execute external commands. Without careful input validation, this enables command injection.
headers.chmod(0o600)
target = config["siteurl"] + "upload/" + quote(args.name or file.name, safe="")
response = Path(directory) / "response.json"
result = subprocess.run([
curl, "--disable", "--silent", "--show-error", "--globoff",
"--proto", "=http,https", "--connect-timeout", "30", "--max-time", "300",
"--header", "@" + str(headers), "--upload-file", str(file),
subprocess module calls execute external commands. Without careful input validation, this enables command injection.
self.log = self.root / "state/openclaw-linx/uploads.json"
def run_upload(self, *options):
return subprocess.run([sys.executable, str(HELPER), *options, "--", str(self.source)],
env=self.env, capture_output=True, text=True, timeout=20)
def install_client(self):
No suspicious patterns detected.