ClawHub
Back to skill

Security audit

Local Websearch 1

Security checks for vulnerabilities and agentic risk

Overview

This is a straightforward SearXNG web-search skill, with the main caveat that users should use an HTTPS SearXNG endpoint because the setup example allows plaintext HTTP.

Install only if you intend your agent to send search queries to the SearXNG service named by SEARXNG_URL. Prefer an HTTPS endpoint, avoid putting credentials in the URL, and treat queries as visible to that SearXNG server and its logs. Also verify the package layout or command path before relying on the skill operationally.

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 (1)

T09 · Insecure Skill Coding Practices

Warning
Location
searxng_search.py:55
Finding

Search Queries Can Be Transmitted over Plaintext HTTP

Content
View full analysis
str: """Get and validate SEARXNG_URL from environment.""" url = os.environ.get("SEARXNG_URL", "").strip() if not url: return "" # Normalize: remove trailing slash, ensure no /search suffix for base url = url.rstrip("/") if url.endswith("/search"): url = url[:-7] return url ``` ### Technical Analysis The Skill must transmit a user-provided query to a SearXNG instance to perform its declared web-search function. That network access is functionally necessary and does not, by itself, exceed minimum privilege. However, the implementation accepts arbitrary URL schemes, and the documentation recommends an `http://` configuration. The search query is included in the request URL through the `q` query-string parameter. When HTTP is used, neither the query nor the returned search results receive transport confidentiality or integrity protection. A network observer, compromised proxy, or other on-path attacker could inspect sensitive search terms or modify the response in transit. URL-based que ...[truncated 1493 chars]
Remediation
View remediation
str: url = os.environ.get("SEARXNG_URL", "").strip().rstrip("/") if not url: return "" parsed = urlparse(url) loopback_hosts = {"localhost", "127.0.0.1", "::1"} if parsed.scheme != "https": if parsed.scheme != "http" or parsed.hostname not in loopback_hosts: raise ValueError( "SEARXNG_URL must use HTTPS; HTTP is allowed only for loopback endpoints" ) if url.endswith("/search"): url = url[:-7] return url ``` ]]>

SkillSpector

Vulnerability Patterns
  • 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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (3)

Tainted flow: 'req' from os.environ.get (line 62, 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 · searxng_search.py (reported line 68)May include surrounding context.

python
})

    try:
        with urllib.request.urlopen(req, timeout=30) as resp:
            data = json.loads(resp.read().decode())
    except urllib.error.HTTPError as e:
        return {"error": f"HTTP {e.code}: {e.reason}", "query": query}

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
91% confidence
Finding

The skill invokes a Python script, requires an environment variable, and performs outbound network access, but it does not declare any explicit tool scope such as permissions or allowed-tools. This creates an authorization and transparency gap: an orchestrator or reviewer cannot easily constrain what the skill is allowed to access, increasing the risk of unintended data exposure or misuse of network capabilities.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger phrases are broad enough to match many ordinary requests such as 'look up', 'what is', or 'google', which can cause the skill to activate in situations where web access was not clearly intended. Over-broad routing increases the chance of unnecessary external queries, inadvertent disclosure of user prompts to the SearXNG instance, and tool overuse.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.