web-search-via-searxng-5ffc92 — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited web-search-via-searxng-5ffc92 (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Use the configured SearXNG MCP server to search the web.
Prefer delegating to the `web-searcher` agent via the Agent tool. The agent handles multi-language parallel search, source credibility verification, cross-validation, iterative deepening, and structured summarization — returning a clean result without polluting your context window with raw JSON. Only call the MCP search tools directly when:
web-searcher agent is unavailable (e.g., no subagent support)When searching directly (without the agent), use these tools:
searchMain search tool. Parameters:
query (required): Search termscategories: Comma-separated category names (e.g., 'general,news,science'). Prefer this over engines — categories leverage multiple engines automaticallylanguage: Language code (e.g., zh, en, ja). Filters results to the specified languagetime_range: day, week, month, yearsafesearch: 0 (off), 1 (moderate), 2 (strict)pageno: Starting page number (default 1)pages: Number of pages to fetch in parallel, 1-5 (default 1). Use 2-3 for comprehensive researchmax_results: Maximum results to return (default 10, max 100)format: compact (default, title/url/content only) or full (includes engines, score, dates)engines: Comma-separated engine names (e.g., google,arxiv,wikipedia). Only use when you need a specific sourceautocompleteGet search query suggestions. Use before searching to discover relevant terms. Best results come from 1-2 meaningful keywords (e.g., "python async"). Single characters return overly broad suggestions; full sentences return none. Makes an external API call to the configured backend (e.g., Bing, Google).
query (required): 1-2 keyword query to autocomplete (e.g., "python async", not "p" or full sentences)engine_infoDiscover available search engines and categories. No parameters. Returns engines grouped by category. Response cached for 5 minutes.
Use this when you need to target specific engines or categories (e.g., "search academic papers" → call engine_info to find science engines, then search with categories=science).
Use categories as the primary filter. Only fall back to engines when categories produce poor results.
| User intent | Category | When to use engines instead |
|---|---|---|
| Current events, breaking news | news | — |
| Academic papers, research | science or scientific publications | arxiv for preprints, pubmed for biomedical |
| Code, libraries, tech docs | it | github for repos, stackoverflow for Q&A |
| Package lookup (npm, pip) | packages | pypi, docker hub by name |
| Images, photos | images | — |
| Videos | videos | youtube for YouTube-specific |
| Maps, locations | map | — |
| Definitions, word meanings | dictionaries or define | — |
| Weather | weather | — |
| Translation | translate | — |
| Shopping, products | shopping | — |
| Torrents, file search | files | — |
| Social media posts | social media | — |
| General knowledge | general (default) | — |
engines when you need a specific source (e.g., arxiv for preprints, github for repos)pages=3 when you need comprehensive resultslanguage when the user writes in a specific languageformat=full when you need to evaluate result quality (scores, engines)autocomplete to refine ambiguous queries before searchingengine_info only when the category table above doesn't cover the use case~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.