Searxng Search
Analysis
This appears to be a straightforward SearXNG search connector, with the main considerations being trusted endpoint use and user-directed setup steps.
Findings (3)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.
cp config.json ~/.config/mcporter/config.json ... npm install -g mcporter
The documented setup uses a global npm-installed dependency and copies a config file into the user's mcporter configuration. This is user-directed and purpose-aligned, but users should be aware of the dependency and config-file impact.
Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.
url = f"{SEARXNG_URL}/search?q={encoded_query}&format=json"
with urllib.request.urlopen(url, timeout=30) as response:The MCP server sends the user's search query to the configured SearXNG endpoint. This is purpose-aligned, but the endpoint receives the search terms.
f" {r.get('content', 'N/A')[:200]}...\n"
...
"text": f"Search results for '{query}':\n\n" + "\n\n".join(formatted_results)The tool returns web result snippets as text to the agent. Web results are untrusted retrieved context and may contain misleading or instruction-like content.
