ClawHub
Back to skill

Security audit

clawhub-skill-remote-agent

Security checks for vulnerabilities and agentic risk

Overview

This skill is a plausible remote-agent bridge, but it can send prompts and an auth token to externally chosen endpoints with weak transport safeguards.

Install only if you control and trust the remote endpoint. Do not use production REMOTE_AGENT_KEY with --url overrides or --insecure mode, and avoid sending secrets, regulated data, internal policy questions, or action-oriented requests unless your organization has approved that remote agent and its data handling.

A.I.G

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (2)

T09 · Insecure Skill Coding Practices

Error
Location
scripts/client.py:12
Finding

Bearer Token Disclosure Through an Unrestricted Endpoint Override

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/client.py:14
Finding

Optional TLS Certificate and Hostname Verification Bypass

Content
View full analysis
Remediation
View remediation

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Taint TrackingDirect Taint Flow, Variable-Mediated Taint Flow, Credential Exfiltration Chain
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (4)

Tainted flow: 'req' from os.getenv (line 69, 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/client.py (reported line 80)May include surrounding context.

python
ctx.verify_mode = ssl.CERT_NONE
            print(json.dumps({"warning": "SSL verification disabled"}, indent=2), file=sys.stderr)
        
        with urllib.request.urlopen(req, context=ctx) as response:
            response_body = response.read().decode("utf-8")
            
            # Try to parse as JSON for pretty printing, otherwise return raw text

Missing User Warnings

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The skill describes delegation to external agents over HTTP but does not clearly warn users that their queries may be transmitted to a third-party or remote service. Because the examples include enterprise knowledge, financial analysis, and pipeline execution, the missing disclosure is more dangerous in context: users may provide sensitive internal data without understanding it will leave the local environment.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding

The skill exposes network and environment-variable capabilities but does not declare any explicit tool scope or permissions boundary. That omission makes the skill harder to review and govern, and can allow an agent to invoke external HTTP requests and use secrets without clear policy constraints or user awareness.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The usage guidance says to forward requests whenever they fall into a specialized domain, which is broad enough to capture sensitive enterprise, financial, or operational queries without meaningful limits. In practice, this can cause over-delegation of confidential prompts to a remote service, increasing the risk of data leakage, unauthorized processing, or unsafe remote actions.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.