searxng-http-mcp@1.2.1
CAUTION· not startedPublisher claims: "SearXNG metasearch engine MCP server" Observed in sandbox: no network, touches nothing outside the project, writes no memory.
Coverage
Not started: dependency install failed; the server itself was not observed.- Tested
- the dependency install step · every file read, file written and host contacted by what ran
- Not tested
- server startup · the tool list · tool behavior
What this means for the label: Caution here covers what ran. The server's tools were not exercised with a real credential, so nothing about their behavior is asserted; the credential summary below says what to do with that.
Before you give it your token
- It will ask for
API_KEYConfiguration, not credentials: CACHE_TTL, SEARXNG_URL- Where it talks to
- no vendor host named in the code
- What the honeypot saw
- no credential files read · no network
- What we could not test
- server startup · the tool list · tool behavior
Connect it with a scoped credential, under the guard. It asks for API_KEY and talks to its vendor. Give the credential to the guard, not the server: the value never enters the server's environment and only travels to the host you bind. What the tools do with a real credential was not observed.
Never paste a credential into a chat, a ticket or a report, including ours. How the guard holds it.
Is [email protected] safe to install?
KillerSkills ran it in the KillerSkills honeypot sandbox (2 scenarios; not started: dependency install failed; the server itself was not observed) on 2026-11-18 and scored it 79/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: Collapsed <details> block carrying instructions; Collapsed <details> block carrying instructions; Collapsed <details> block carrying instructions. Summary: no network, touches nothing outside the project, writes no memory.
Behavior trace
Honeypot run · 0 s · 2 scenarios · 0 deviationsStatic findings
<summary><img src="assets/icons/claude.svg" height="16" alt="Claude"> <b>Claude Desktop</b></summary> **Server mode** — edit `~/Library/Application Support/ClauFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.<summary><img src="assets/icons/cursor.svg" height="16" alt="Cursor"> <b>Cursor</b></summary> **Server mode** — edit `.cursor/mcp.json`: ```json { "mcpServers":Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.<summary><img src="assets/icons/githubcopilot.svg" height="16" alt="Copilot"> <b>VS Code Copilot</b></summary> **Server mode** — add to `.vscode/mcp.json`: ```jFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.You MUST follow these steps in order. Do NOT skip any step.Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.<summary><img src="assets/icons/cursor.svg" height="16" alt="Cursor"> <b>Cursor</b></summary> **服务器模式** —— 编辑 `.cursor/mcp.json`: ```json { "mcpServers": { "seaFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.<summary>为什么工具更少?</summary> MCP 的设计哲学是组合 — 客户端连接多个专用 server,每个做好一件事。一些竞品把 URL 抓取、渲染提取、多查询并行或研究模式塞进搜索 server。我们有意将工具面控制在三个(搜索、自动补全、引擎发现): - **URL 抓取是独立关注点。** MCPFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.<summary><img src="assets/icons/codex.svg" height="16" alt="Codex"> <b>Codex</b></summary> **Server mode** — add to `~/.codex/config.toml`: ```toml [mcp_serversFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.<summary><img src="assets/icons/claude.svg" height="16" alt="Claude"> <b>Claude Code</b></summary> **Server mode**: ```bash claude mcp add --transport http --heFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.<summary><img src="assets/icons/codex.svg" height="16" alt="Codex"> <b>Codex</b></summary> **服务器模式** —— 添加到 `~/.codex/config.toml`: ```toml [mcp_servers.searxngFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.<summary><img src="assets/icons/claude.svg" height="16" alt="Claude"> <b>Claude Desktop</b></summary> **服务器模式** —— 编辑 `~/Library/Application Support/Claude/clauFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.<summary><img src="assets/icons/claude.svg" height="16" alt="Claude"> <b>Claude Code</b></summary> **服务器模式**: ```bash claude mcp add --transport http --header "Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.@patch("mcp_server.tools.httpx.AsyncClient")Fix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.@patch("mcp_server.proxy.httpx.AsyncClient")Fix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.@patch("mcp_server.tools.httpx.AsyncClient")Fix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.with patch("mcp_server.proxy.httpx.AsyncClient") as mock_client_cls:Fix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.rm /tmp/requirements.pip.txt /tmp/requirements.docker.txt && \Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.@patch("mcp_server.proxy.httpx.AsyncClient")Fix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.RUN chmod +x /usr/local/searxng/custom-entrypoint.shFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.You MUST follow these steps in order. Do NOT skip any step.Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.@patch("mcp_server.proxy.httpx.AsyncClient")Fix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.RUN chmod +x /usr/local/searxng/healthcheck.shFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.@patch("mcp_server.tools.httpx.AsyncClient")Fix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.@patch("mcp_server.proxy.httpx.AsyncClient")Fix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.docker run --rm \Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.with patch("mcp_server.proxy.httpx.AsyncClient") as mock_client_cls:Fix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.@patch("mcp_server.tools.httpx.AsyncClient")Fix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.@patch("mcp_server.proxy.httpx.AsyncClient")Fix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.with patch("mcp_server.proxy.httpx.AsyncClient") as mock_client_cls:Fix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.with patch("mcp_server.proxy.httpx.AsyncClient") as mock_client_cls:Fix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.exec python -m mcp_server.main --httpFix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.@patch("mcp_server.tools.httpx.AsyncClient")Fix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.You MUST follow these steps in order. Do NOT skip any step.Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.docker run --rm \Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.Cost to the agent
Tokens estimated at four characters each. The tool list is re-read by the agent on every turn, so it is the number to compare between servers that do the same job: ranked across the index.
Known vulnerabilities
No advisories in OSV for searxng-http-mcp@1.2.1 as of 21 min ago. This is the package itself, not its dependencies.
What would make this Verified
1 point to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%- Unclaimed publisher≈ +8 on the totalClaim pypi: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
- Hidden instructions≈ +6 on the totalRemove HTML comments, zero-width characters and white-on-white text that carry instructions. Everything the agent is meant to read should be visible to the person reading the file.
- MCP server using network transport (SSE/HTTP)≈ +2 on the totalNamespace tool names and keep tool descriptions literal; no instructions to the agent inside them.
- Destructive file system command detected≈ +2 on the totalBuild commands from fixed arguments; do not interpolate untrusted strings into a shell.
- File permission modification command≈ +2 on the totalRequest the narrowest permission that works; replace wildcards with the exact paths or hosts.