Back to skill

Security audit

agent-health-monitor

Security checks for vulnerabilities and agentic risk

Overview

ClawScan marks this skill suspicious with medium confidence, mainly because installation can pull unpinned dependencies and the scan could not resolve several referenced paths.

Treat this as a review decision rather than a finding of malicious intent. Before installing, consider whether you need to provide it credentials or broad permissions, and whether you can accept the risk of unpinned dependencies. The scan inspected all listed components but was marked partial because some references could not be resolved, so its evidence is incomplete.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • Taint TrackingDirect Taint Flow, Variable-Mediated Taint Flow, Credential Exfiltration Chain
Findings (21)

Tainted flow: 'req' from os.getenv (line 88, credential/environment) → urllib.request.urlopen (network output)

Critical
Category
Data Flow
Confidence
90% confidence
Finding

Credentials or environment variables flow to a network sink. This is a high-confidence indicator of credential exfiltration.

Content

Scanner excerpt · scripts/alert_bot.py (reported line 89)May include surrounding context.

python
url = f"https://api.telegram.org/bot{self.token}/sendMessage"
                data = json.dumps({"chat_id": self.chat_id, "text": message, "parse_mode": "Markdown"}).encode()
                req = urllib.request.Request(url, data=data, headers={"Content-Type": "application/json"})
                with urllib.request.urlopen(req, timeout=10):
                    pass
            return True
        except Exception as e:

Tainted flow: 'req' from os.getenv (line 105, credential/environment) → urllib.request.urlopen (network output)

Critical
Category
Data Flow
Confidence
90% confidence
Finding

Credentials or environment variables flow to a network sink. This is a high-confidence indicator of credential exfiltration.

Content

Scanner excerpt · scripts/alert_bot.py (reported line 106)May include surrounding context.

python
import urllib.request
            data = json.dumps({"content": message}).encode()
            req = urllib.request.Request(self.discord_webhook, data=data, headers={"Content-Type": "application/json"})
            with urllib.request.urlopen(req, timeout=10):
                pass
            return True
        except Exception as e:

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · scripts/alert_bot.py (reported line 19)May include surrounding context.

python
Why the credential-bearing requests below are NOT exfiltration
(ClawHub audit TT3/E1): sending a Telegram message REQUIRES the bot
token — Telegram's own Bot API design puts the token in the request
URL (https://api.telegram.org/bot<token>/sendMessage), and there is no
token-free way to use it. Likewise a Discord webhook URL IS the
credential by design, and is allow-listed to Discord's own webhook
hosts only (see _valid_discord_webhook). Both destinations are the

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · references/alert-config.md (reported line 6)May include surrounding context.

md
## Telegram Setup
1. Create bot via @BotFather
2. Get token: `TELEGRAM_BOT_TOKEN=123456:ABC-DEF...`
3. Get chat ID: message bot, visit `https://api.telegram.org/bot<token>/getUpdates`
   (Telegram's official Bot API endpoint — the token is in the URL by
   Telegram's own design; see "How credentials are handled" in SKILL.md.)

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · scripts/alert_bot.py (reported line 86)May include surrounding context.

python
await self._bot.send_message(chat_id=self.chat_id, text=message, parse_mode="Markdown")
            else:
                import urllib.request
                url = f"https://api.telegram.org/bot{self.token}/sendMessage"
                data = json.dumps({"chat_id": self.chat_id, "text": message, "parse_mode": "Markdown"}).encode()
                req = urllib.request.Request(url, data=data, headers={"Content-Type": "application/json"})
                with urllib.request.urlopen(req, timeout=10):

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · scripts/api_key_monitor.py (reported line 81)May include surrounding context.

python
"""Test Groq API key."""
    try:
        req = urllib.request.Request(
            "https://api.groq.com/openai/v1/models",
            headers={"Authorization": f"Bearer {key}"}
        )
        resp = urllib.request.urlopen(req, timeout=10)

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · scripts/api_key_monitor.py (reported line 99)May include surrounding context.

python
"""Test Mistral API key."""
    try:
        req = urllib.request.Request(
            "https://api.mistral.ai/v1/models",
            headers={"Authorization": f"Bearer {key}"}
        )
        resp = urllib.request.urlopen(req, timeout=10)

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · scripts/api_key_monitor.py (reported line 133)May include surrounding context.

python
"""Test GitHub API key."""
    try:
        req = urllib.request.Request(
            "https://api.github.com/user",
            headers={"Authorization": f"token {key}", "User-Agent": "Agent"}
        )
        resp = urllib.request.urlopen(req, timeout=10)

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · scripts/api_key_monitor.py (reported line 150)May include surrounding context.

python
"""Test Dune API key."""
    try:
        req = urllib.request.Request(
            "https://api.dune.com/api/v1/query/1/results",
            headers={"x-dune-api-key": key}
        )
        resp = urllib.request.urlopen(req, timeout=10)

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · scripts/healthcheck.py (reported line 99)May include surrounding context.

python
return check(name, False, str(e)[:40])

def check_groq():
    return _probe_reachable("Groq", "https://api.groq.com/openai/v1/models", "GROQ_API_KEY")

def check_mistral():
    return _probe_reachable("Mistral", "https://api.mistral.ai/v1/models", "MISTRAL_API_KEY")

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · scripts/healthcheck.py (reported line 102)May include surrounding context.

python
return _probe_reachable("Groq", "https://api.groq.com/openai/v1/models", "GROQ_API_KEY")

def check_mistral():
    return _probe_reachable("Mistral", "https://api.mistral.ai/v1/models", "MISTRAL_API_KEY")

def check_openrouter():
    # /models is public on OpenRouter — reachable without any credential.

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · scripts/healthcheck.py (reported line 109)May include surrounding context.

python
return _probe_reachable("OpenRouter", "https://openrouter.ai/api/v1/models", "OPENROUTER_API_KEY")

def check_dune():
    return _probe_reachable("Dune", "https://api.dune.com/api/v1/status", "DUNE_API_KEY")

def check_github():
    # api.github.com root answers 200 without auth; /user would need the token.

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · scripts/healthcheck.py (reported line 113)May include surrounding context.

python
def check_github():
    # api.github.com root answers 200 without auth; /user would need the token.
    return _probe_reachable("GitHub", "https://api.github.com/", "GITHUB_TOKEN")

def check_bankr():
    key = os.getenv("BANKR_API_KEY")

subprocess module call

Medium
Category
Dangerous Code Execution
Confidence
70% confidence
Finding

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

Content

Scanner excerpt · scripts/master_monitor.py (reported line 48)May include surrounding context.

python
if not quiet:
        print("\n--- Healthcheck ---")
    try:
        result = subprocess.run(
            [sys.executable, os.path.join(SCRIPT_DIR, "healthcheck.py"), "--quiet"],
            capture_output=True, text=True, timeout=60
        )

subprocess module call

Medium
Category
Dangerous Code Execution
Confidence
70% confidence
Finding

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

Content

Scanner excerpt · scripts/master_monitor.py (reported line 66)May include surrounding context.

python
if not quiet:
        print("\n--- Balance Monitor ---")
    try:
        result = subprocess.run(
            [sys.executable, os.path.join(SCRIPT_DIR, "balance-monitor.py")],
            capture_output=True, text=True, timeout=30
        )

Permission 'file_read' is declared but no corresponding code capability (file_read) was detected.

Low
Category
MCP Least Privilege
Confidence
65% confidence
Finding

Declared permissions with no matching code capability may indicate removed functionality or pre-staging for future abuse.

Content

No source excerpt is available for this finding.

Excessive Permissions

Low
Category
Privilege Escalation
Confidence
60% confidence
Finding

Skill requests more permissions than appear necessary for its stated functionality. Review if elevated access is justified.

Content

Scanner excerpt · SKILL.md (reported line 190)May include surrounding context.

md
### v1.0.2 — 2026-11-19 (rescan follow-up)

- Declared the full tool scope in the frontmatter (`permissions: [env, file_read, file_write, network, shell]`) — the v1.0.1 list omitted `env` and `file_write`, which the rescan flagged as HIGH LP1. No code change; behavior is identical to v1.0.1.

### v1.0.1 — 2026-11-18 (security audit remediation)

Unpinned Dependencies

Low
Category
Supply Chain
Confidence
97% confidence
Finding

The web3 dependency has no version or hash constraint, allowing installations to resolve to different, unreviewed releases and transitive dependencies. This is a supply-chain and reproducibility weakness, especially because installed code runs in a monitoring environment containing credentials; it does not prove that any particular release is compromised.

Content

Scanner excerpt · requirements.txt (reported line 8)May include surrounding context.

text
# healthcheck.py uses web3 and eth-account for the Alchemy RPC and
# executor-key checks. Without web3 the balance check cannot run at all
# (master_monitor.py reports that as a failure; it never silently passes).
web3
eth-account

# Optional: alert_bot.py prefers python-telegram-bot when installed and

Unverifiable Dependency: web3 has 2 known advisory(ies) (CVE-2026-111171 (web3.py: SSRF via CCIP Read (EIP-3668) OffchainLookup URL handling); CVE-2026-111171 (web3.py: SSRF via CCIP Read (EIP-3668) OffchainLookup URL handling)), but the manifest does not pin a version, so it is unknown whether the installed release is affected

Low
Category
Supply Chain
Confidence
40% confidence
Finding

OSV returned advisories for this package, but its resolved version is unknown. A package-name lookup does not establish whether the installed release is affected by those advisories.

Content

No source excerpt is available for this finding.

Unpinned Dependencies

Low
Category
Supply Chain
Confidence
97% confidence
Finding

The eth-account dependency is unconstrained, so installation can select releases that have not been reviewed or tested with this skill. The context increases the potential consequences because the package may handle the optional executor private key locally, although no compromised package is demonstrated.

Content

Scanner excerpt · requirements.txt (reported line 9)May include surrounding context.

text
# executor-key checks. Without web3 the balance check cannot run at all
# (master_monitor.py reports that as a failure; it never silently passes).
web3
eth-account

# Optional: alert_bot.py prefers python-telegram-bot when installed and
# falls back to plain HTTPS otherwise. Uncomment to use it:

Unverifiable Dependency: eth-account has 2 known advisory(ies) (CVE-2022-1930 (Regular expression denial of service in eth-account); CVE-2022-1930 (Regular expression denial of service in eth-account)), but the manifest does not pin a version, so it is unknown whether the installed release is affected

Low
Category
Supply Chain
Confidence
40% confidence
Finding

OSV returned advisories for this package, but its resolved version is unknown. A package-name lookup does not establish whether the installed release is affected by those advisories.

Content

No source excerpt is available for this finding.