searxng

SAFE max severity MEDIUMSkillSpector Do not install 100 registry clawhub.ai last scanned 2026-02-07 1 scan

Verdict safe, but 2 findings (2 medium), mainly data exfiltration, command injection. Most recent scan 2026-02-07.

The searxng skill is a privacy-focused web search tool that interfaces with a local SearXNG instance. The primary security concerns are: (1) unvalidated external network requests to user-controlled URLs, allowing potential data exfiltration if misconfigured, and (2) disabled SSL certificate verification creating MITM vulnerability. The skill does not contain prompt injection, credential theft, or malicious code execution. However, the lack of URL validation and disabled SSL verification represent moderate security risks. The skill's behavior matches its description, and there are no deceptive elements. With proper URL validation and SSL verification, this would be a safe skill for local use.
Scanner's overall assessment (LLM analyzer)

Findings2

Critical0

High0

Medium2

Low0

Analyzersstatic llm meta

Second scanner: NVIDIA SkillSpector

VerdictDo not install 100

Score100 / 100

Issues11

CategoriesTool Misuse Trigger Abuse Rogue Agent

Version2.12.0 (static)

The two scanners disagree: Cisco found nothing and SkillSpector says do not install. One of them has a blind spot here; read the skill before trusting either. How the two scanners compare across the whole corpus: Two scanners, one corpus.

What this skill does

Privacy-respecting metasearch using your local SearXNG instance. Search the web, images, news, and more without external API dependencies.

Package: 6 files, 19 KB
SKILL.md: 179 words, 8 sections
Code: Python (1)
Requires: bins: python3
Version: 1.0.1
Homepage: https://searxng.org

SKILL.md covers: SearXNG Search · Commands · Web Search · Category Search · Advanced Options · Configuration · Features · API

Findings

MEDIUM Unvalidated External Network Requests to User-Controlled URL Data Exfiltrationllm

The skill makes HTTP requests to a user-controlled SearXNG instance URL (SEARXNG_URL environment variable) without validation or restrictions. While the default is localhost, users can configure this to point to any external server. The skill sends search queries and receives responses from this endpoint without verifying the destination is trustworthy. An attacker could configure SEARXNG_URL to point to a malicious server that logs queries or returns crafted responses to exploit the agent.

scripts/searxng.py

SEARXNG_URL = os.getenv("SEARXNG_URL", "http://localhost:8080")
...
response = httpx.get(f"{SEARXNG_URL}/search", params=params, verify=False, timeout=30.0)

Remediation: 1. Validate SEARXNG_URL format (must be http://localhost or https://localhost with optional port). 2. Reject external URLs unless explicitly allowlisted. 3. Add URL validation function to ensure only local instances are used. 4. Document security implications of pointing to external instances.

MEDIUM SSL Certificate Verification Disabled Command Injectionllm

The script explicitly disables SSL certificate verification (verify=False) and suppresses SSL warnings. This makes the connection vulnerable to man-in-the-middle attacks if users configure an HTTPS SearXNG instance. While intended for local self-signed certificates, this creates a security risk when users point to external HTTPS instances.

scripts/searxng.py

warnings.filterwarnings('ignore', message='Unverified HTTPS request')
...
response = httpx.get(f"{SEARXNG_URL}/search", params=params, verify=False, timeout=30.0)

Remediation: 1. Remove verify=False and allow proper SSL verification. 2. If self-signed certificates are needed, provide documentation on proper certificate installation. 3. Add a configuration option for certificate verification rather than disabling it globally. 4. Remove the SSL warning suppression.

What is in the package

File	Kind	Size
`CHANGELOG.md`	doc	1.3 KB
`PUBLISH.md`	doc	2.5 KB
`PUBLISHING_CHECKLIST.md`	doc	3.2 KB
`README.md`	doc	4.2 KB
`SKILL.md`	doc	1.9 KB
`scripts/searxng.py`	script	6.0 KB

External hosts referenced

your-searxng-instance.com clawdhub.com docs.searxng.org keepachangelog.com searxng.org semver.org

Hostnames that appear in the skill's files. Referencing a host is not a finding; compare against the findings above and the skill's stated purpose.

Environment variables referenced

SEARXNG_URL

Package contents recorded 2026-10-22. The scan above ran against the version current at scan time; the two can differ if the skill has since been updated.

What we checked

Static analyzer. Pattern rules over SKILL.md and bundled files: known-bad commands, exfiltration URLs, secrets, obfuscation.
Llm analyzer. A language model reads the skill and reports intent-level issues the rules cannot express.
Meta analyzer. A second model pass that validates, deduplicates and re-scores the other analyzers' findings.

What we did not check

Runtime behaviour. The skill was not executed; findings come from reading its files.
Anything fetched at install or run time, including remote scripts, package dependencies, and URLs the skill tells the agent to open.
Content the analyzers cannot read: images, binaries, and some non-English text.
Known scanner bypass classes. See the methodology and limitations page.

Sources

Repository: https://clawhub.ai/skills/searxng
Raw scan result: searxng-scan.json
Badge for your README: ![skillscan.io](https://skillscan.io/badge/clawhub-ai-searxng.svg)
Request a re-scan