SkillTotal

Is headroomlabs-ai/headroom safe?

Some risk - review before installing
Notable — review in context (capabilities are not malware):
  • Python shell/command execution
  • Python dynamic code execution
  • Node.js shell/command execution

What to do: Read the findings below before installing: each one opens the exact line of code it was found on.

headroomlabs-ai/headroom is a PyPI package analyzed by SkillTotal's deterministic static scanner. The scan found no malicious indicators, though 2 risky constructs are reported for review. It can: delegated authentication, dynamic code execution, filesystem read, filesystem write, mcp tools detected, network egress and shell execution — capabilities are what the code can do, not a verdict on intent. Risk score 30/100 (medium).

headroom-ai 0.40.0

python_package · https://github.com/headroomlabs-ai/headroom
MEDIUM
30
/ 100 risk score
Snapshot · scanned Oct 8, 2026 · headroom-ai@0.40.0 · engine 0.56.4 / ruleset 62

Automated static-analysis result. It can contain false positives and false negatives, and is not a claim about the intent of headroomlabs-ai/headroom's authors. Report a false positive.

Capabilities — what this component can do (not a risk score):
delegated authenticationdynamic code executionfilesystem readfilesystem writemcp tools detectednetwork egressshell execution

Behavioral traits

How this component maps to the CSA agentic threat model. Descriptive — it never affects the risk score.

Tool surface
Tool Usage
Execution authority
Tool Access Control / Direct Tool Access
Filesystem reach
Tool Execution Context
Network egress
Interaction & Communication / Direct Communication
Network exposure
Interaction & Communication
Embedded credential access
Tool Execution Context / Agent Service Identity
Delegated authentication
Tool Execution Context / User Delegated Credentials

Findings (14)

HIGHPython dynamic code executionST-DYN-PY

The code turns strings into live code at runtime (eval / new Function / exec).

compile(code, "<headroom-compressed>", "exec")
return ExportWrapper(core).eval()

Why it matters: If those strings aren't fixed and trusted, they become a way to run arbitrary code.

Fix: Avoid evaluating dynamically constructed code; if unavoidable, ensure the input is a trusted constant and never derived from external data.

HIGHEmbedded secret / credentialnot scoredST-SECRET-EMBEDDED

A hardcoded credential (API key, token, or private key) is shipped in the code.

"Bearer sk-a…[redacted, 49 chars]",
"Bearer sk-a…[redacted, 40 chars]",

Why it matters: Anyone who gets the package gets the secret — rotate it and load secrets at runtime instead.

Fix: Remove the secret from the code, rotate it immediately, and load credentials from the environment or a secrets manager at runtime.

HIGHDefense-evasion command idiomST-SHELL-EVASION

A command uses a known defense-evasion idiom: PowerShell execution-policy bypass / encoded command / hidden window, macOS code-signing bypass, or launching a payload from a world-writable temp directory. These are hallmarks of droppers and rarely appear in legitimate code. (3 occurrence(s) shown as evidence).

$cmdWrapper = ([string][char]64) + "echo off`r`npowershell -NoLogo -NoProfile -ExecutionPolicy Bypass -File ""%~dp0headroom.ps1"" %*`r`n"

Fix: Verify why the component bypasses execution policy / code signing or runs from a temp directory; these patterns are characteristic of malware staging.

HIGHNode.js shell/command executionST-SHELL-NODE

The component can run operating-system commands or spawn processes.

import { spawn } from "node:child_process";
import { spawnSync } from "node:child_process";
const child = spawn(spec.command, spec.args, {
const result = spawnSync("pyenv", ["which", "headroom"], {
const result = spawnSync(command, args, {
const result = spawnSync("npm", ["prefix", "-g"], {
import { spawnSync } from "node:child_process";
const result = spawnSync(command, args, {
import { spawnSync } from "node:child_process";
const result = spawnSync("tar", ["-xzf", tarballPath, "-C", workdir], {
return spawnSync("cmd.exe", ["/d", "/s", "/c", "npm.cmd", ...args], {
const result = spawnSync(process.execPath, ["--input-type=module", "-e", smoke], {

Why it matters: Powerful and often legitimate — confirm the commands aren't built from untrusted input.

Fix: Confirm the command and its arguments are fully controlled and not derived from untrusted input; prefer execFile with an argument array.

HIGHPython shell/command executionST-SHELL-PY

The component can run operating-system commands or spawn processes.

return _sp.run(*args, **kwargs)
return _sp.Popen(*args, **kwargs)
result = subprocess.run(command)
completed = subprocess.run(cmd, check=False)
result = subprocess.run(argv)  # noqa: S603
subprocess.Popen(handoff_argv)  # noqa: S603 - fixed allowlist plus helper argv
proc = subprocess.Popen(cmd, **popen_kwargs)
proc = subprocess.Popen(cmd, **popen_kwargs)
subprocess.run(
                ["taskkill", "/F", "/T", "/PID", str(proc.pid)],
                capture_output=True,
                timeout=10,
                check=False,
            )
proc = subprocess.Popen(
            [
                "uvx",
                # PyPI (prebuilt wheels), not the git source that fails to build
                # under proot-based filesystems (#2871).
                "--from", …
subprocess.run(
                ["taskkill", "/F", "/T", "/PID", str(pid)],
                capture_output=True,
                timeout=10,
                check=False,
            )
result = subprocess.run([binary, *args], env=env)
result = subprocess.run([claude_bin, *claude_args], env=env)
proc = subprocess.Popen(
            [sys.executable, "-m", "headroom.proxy.server", "--port", str(port)],
            stdout=subprocess.PIPE,
            stderr=subprocess.PIPE,
        )
subprocess.run(command, check=True)
subprocess.run(command, check=True)
proc = subprocess.Popen(command, env=env, stdout=log_file, stderr=log_file)
proc = subprocess.Popen(command, **kwargs)
subprocess.run(docker_cmd, check=True)
subprocess.Popen(command, **popen_kwargs)
subprocess.run(
                ["schtasks", "/Create", "/TN", name, "/XML", tmp.name, "/F"],
                check=True,
            )
query = subprocess.run(
                ["schtasks", "/Query", "/TN", name, "/XML"],
                check=True,
                capture_output=True,
            )
subprocess.run(["systemctl", *flags, "daemon-reload"], check=True)
subprocess.run(["systemctl", *flags, "enable", manifest.service_name], check=True)
subprocess.run(["systemctl", *flags, "restart", manifest.service_name], check=True)

Why it matters: Powerful and often legitimate — confirm the commands aren't built from untrusted input.

Fix: Confirm the command and its arguments are fully controlled and not derived from untrusted input; avoid shell=True.

MEDIUMServer bound to all network interfacesST-EXPOSE-BIND

A server is bound to all network interfaces (0.0.0.0), not just your own machine.

CONTAINER_BIND_HOST = "0.0.0.0"  # noqa: S104 — container-internal bind, published only on 127.0.0.1
body = signature.split("::", 1)[-1]
With the default ``--host 0.0.0.0`` Docker bind, leaving it open would
``--host 0.0.0.0`` Docker bind, an unauthenticated POST from any

Why it matters: Without authentication, other hosts on the network can reach it.

Fix: Bind to 127.0.0.1 for local-only use, or require authentication and restrict access if remote exposure is intended.

MEDIUMNode.js filesystem readST-FS-NODE-READ

The component reads files from disk.

await fs.readFile(path.join(rootDir, "package.json"), "utf8"),

Why it matters: Usually legitimate, but worth confirming it can't be steered into reading sensitive files.

Fix: Confirm which files are read and that paths cannot be influenced by untrusted input to reach sensitive locations.

MEDIUMNode.js filesystem write/deleteST-FS-NODE-WRITE

The component writes or deletes files on disk.

Why it matters: Usually legitimate, but worth confirming the paths can't be controlled by untrusted input.

Fix: Confirm which files are written/deleted and that paths cannot be influenced by untrusted input.

MEDIUMPython filesystem readST-FS-PY-READ

The component reads files from disk.

with open(f"/proc/{pid}/stat", "rb") as fh:
with open(path, encoding="utf-8") as f:
with open(path, encoding="utf-8") as f:
with open(path, encoding="utf-8") as f:
rows = [json.loads(x) for x in open(os.environ["HEADROOM_CACHE_TTL_OBS_PATH"])]
for line_number, line in enumerate(capture_path.read_text(encoding="utf-8").splitlines(), 1):
with open(SHARED_STATS_FILE) as f:
content = safe_decode_for_logging(path.read_bytes())
payload = json.loads(path.read_text(encoding="utf-8"))
payload = json.loads(path.read_text(encoding="utf-8")) if path.exists() else {}
payload = json.loads(path.read_text(encoding="utf-8"))
payload = json.loads(path.read_text(encoding="utf-8"))
payload = json.loads(settings_path.read_text(encoding="utf-8"))
text = candidate.read_text(encoding="utf-8", errors="replace")
payload = json.loads(savings_file.read_text(encoding="utf-8"))
content = path.read_text(encoding="utf-8").strip()
content = path.read_text(encoding="utf-8") if path.exists() else ""
content = path.read_text(encoding="utf-8") if path.exists() else ""
payload = json.loads(path.read_text(encoding="utf-8")) if path.exists() else None
text = path.read_text(encoding="utf-8", errors="replace") if path.exists() else ""
record = _json.loads((ws / _VERBOSITY_PIN_FILE).read_text(encoding="utf-8"))
with open(MCP_CONFIG_PATH, encoding="utf-8") as f:
with open(path) as f:
content = fsutil.read_text(file_path)
payload = json.loads(marker.read_text(encoding="utf-8"))

Why it matters: Usually legitimate, but worth confirming it can't be steered into reading sensitive files.

Fix: Confirm which files are read and that paths cannot be influenced by untrusted input to reach sensitive locations.

MEDIUMPython filesystem write/deleteST-FS-PY-WRITE

The component writes or deletes files on disk.

args.output.write_text(json.dumps(payload, ensure_ascii=False) + "\n", encoding="utf-8")
shutil.copyfileobj(src, out)
shutil.copyfileobj(gz, out)
shutil.copy2(archive, dest)
shutil.copyfileobj(extracted, out)
shutil.copyfileobj(src, out)
shutil.move(str(staging), tmp_final)
with open(tmp, "w", encoding="utf-8") as f:
with open(path, "a", encoding="utf-8") as f:
with open(os.environ["HEADROOM_CACHE_TTL_LEARNED_PATH"], "w") as f:
with open(SHARED_STATS_FILE, "a") as f:
with open(SHARED_STATS_FILE, "w") as f:
(logs_dir / "proxy.log").write_text("\n".join(perf_lines) + "\n", encoding="utf-8")
eval_path.write_text(
        json.dumps(
            {
                "totals": {
                    "cases": 3,
                    "passed": 3,
                    "accuracy_rate": 1.0,
                    "tokens_original": 3000, …
markdown_output.write_text(markdown, encoding="utf-8")
json_output.write_text(json.dumps(diff.to_dict(), indent=2), encoding="utf-8")
json_output.write_text(json_module.dumps(report.to_dict(), indent=2), encoding="utf-8")
json_output.write_text(json_module.dumps(report.to_dict(), indent=2), encoding="utf-8")
path.write_text(json.dumps(payload, indent=2) + "\n", encoding="utf-8")
path.write_text(content, encoding="utf-8")
path.write_text(content, encoding="utf-8")
with open(os.devnull, "w", encoding="utf-8") as devnull:
path.write_text(_json.dumps(record), encoding="utf-8", newline="\n")
with open(MCP_CONFIG_PATH, "w", encoding="utf-8") as f:
fsutil.write_text(output_path, json_output)

Why it matters: Usually legitimate, but worth confirming the paths can't be controlled by untrusted input.

Fix: Confirm which files are written/deleted and that paths cannot be influenced by untrusted input.

MEDIUMNode.js network egressST-NET-NODE

The component makes outbound network requests.

async fetch(request, env, ctx) {
fetch(env.METRICS_OTLP_URL, {
const res = await $.http.fetch(`${live.url}/stats?cached=1`)
const response = await fetch(`${origin}${path}`, {
const resp = await fetch(url, {
response = await fetch(url, {

Why it matters: Usually legitimate, but confirm the destinations are expected and no sensitive data leaves.

Fix: Confirm the destination hosts are expected and that no sensitive data is sent off-host.

MEDIUMPython network egressST-NET-PY

The component makes outbound network requests.

import urllib.request
req = urllib.request.Request(final_url, headers={"User-Agent": "headroom-binaries/1"})
resp = urllib.request.urlopen(req, timeout=60, context=ctx)  # noqa: S310 (https)
resp = urllib.request.urlopen(req, timeout=60)  # noqa: S310 (https)
url_path = urllib.parse.urlparse(url).path
self._http_client = httpx.AsyncClient(timeout=15.0)
async with httpx.AsyncClient(timeout=5.0) as client:
self._http_client = httpx.AsyncClient(timeout=15.0)
with urllib.request.urlopen(url, timeout=2) as response:
request = urllib.request.Request(
        f"http://127.0.0.1:{resolved_port}/admin/runtime-env",
        data=_json.dumps(overrides).encode("utf-8"),
        method="POST",
        headers={"Content-Type": "application/json"},
    )
with urllib.request.urlopen(request, timeout=2) as response:
response = httpx.get(f"{proxy_url}/health", timeout=2.0)
with urllib.request.urlopen(url, timeout=2) as response:
parsed = urllib.parse.urlparse(raw)
request = urllib.request.Request(
        f"http://127.0.0.1:{port}/admin/runtime-env",
        data=json.dumps(payload).encode("utf-8"),
        method="POST",
        headers={"Content-Type": "application/json"},
    )
with urllib.request.urlopen(request, timeout=2) as response:
parsed = urllib.parse.urlsplit(base_url)
from urllib import error as urllib_error

Why it matters: Usually legitimate, but confirm the destinations are expected and no sensitive data leaves.

Fix: Confirm the destination hosts are expected and that no sensitive data is sent off-host.

LOWDelegated authentication (OAuth 2.0 / OIDC)ST-AUTH-DELEGATED

An OAuth 2.0 / OpenID Connect delegated-authentication flow was detected (authorization-code / refresh-token / token-exchange grant, an OIDC authorize/discovery endpoint or id_token, or a delegation library). Tools authenticate with the end user's delegated, scoped credentials rather than a long-lived embedded service credential. (13 occurrence(s) shown as evidence).

return _id_token_carries_chatgpt_account(tokens.get("id_token"))
"""Whether an ``id_token`` carries the ChatGPT account claim (#3206).
the ``id_token`` claims. Those configs then read as API-key mode, so
of it. An API-key user has no ChatGPT id_token, so this cannot resurrect

Fix: Delegated auth is a lower-blast-radius execution context than an embedded static credential. Confirm the requested scopes are minimal and that tokens are never logged or forwarded off-host.

LOWMCP tool surface detectedST-MCP-DETECTED

An MCP tool surface (manifest or tool definitions) was found.

Why it matters: Just context — review which tools it offers and their permissions.

Fix: Review the declared MCP tools and their permissions.

How attackers abuse these capabilities

Interactive labs on the attack class behind the rules above. They show the technique, not anything found in headroomlabs-ai/headroom.

Check your own component

Run the same evidence-backed scan on any MCP server, agent skill, or package.

Scan your own component

How we determine this: deterministic static analysis (regex + AST), evidence-anchored, no code execution. Methodology →