{"_id":"llmaudit-mcp","_rev":"2-4c9d202b868712e4d31aaa87b6d4f0df","name":"llmaudit-mcp","dist-tags":{"latest":"0.1.0"},"versions":{"0.1.0":{"name":"llmaudit-mcp","version":"0.1.0","keywords":["mcp","model-context-protocol","mcp-server","ai-visibility","geo","generative-engine-optimization","brand-monitoring","llm","chatgpt","claude"],"author":{"name":"Nahuel Soria"},"license":"MIT","_id":"llmaudit-mcp@0.1.0","maintainers":[{"name":"nahhuelsoria","email":"jorgenahuelsoria@gmail.com"}],"homepage":"https://llmaudit.app/developers","bugs":{"url":"https://github.com/nahuelsoria/llmaudit-mcp/issues"},"bin":{"llmaudit-mcp":"bin/llmaudit-mcp.js"},"dist":{"shasum":"3d0a27bad777df4e705e63b01216f5366b8cffde","tarball":"https://registry.npmjs.org/llmaudit-mcp/-/llmaudit-mcp-0.1.0.tgz","fileCount":4,"integrity":"sha512-nblpE9zmBtP/rtn7QUBLUZ0fcaQdJF9p3MZxbqIAWmN61m3NgEuG3qrcjyIwL6CcCTtx2VJeMAqilAY6eyzdlg==","signatures":[{"sig":"MEUCID4kY27M1SXybfSzSESkR+VaGZMe/ZWdRM2MApKbNOj+AiEAu9Er7jpuDkj9lMCif9I08K/n/WkP5Myr+I5Yk9yOs3M=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":6410},"type":"module","engines":{"node":">=18"},"gitHead":"13d78d48480944b44a5c25f66d3cfdf3975c1edc","mcpName":"app.llmaudit/llmaudit","_npmUser":{"name":"nahhuelsoria","email":"jorgenahuelsoria@gmail.com"},"repository":{"url":"git+https://github.com/nahuelsoria/llmaudit-mcp.git","type":"git","directory":"npm"},"_npmVersion":"10.9.8","description":"MCP server for llmaudit.app: measure whether AI assistants actually recommend a brand. stdio bridge to the hosted server, for clients that launch local commands.","directories":{},"_nodeVersion":"22.23.1","dependencies":{"mcp-remote":"0.8.1"},"_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/llmaudit-mcp_0.1.0_1787926638884_0.9369317216022888","host":"s3://npm-registry-packages-npm-production"},"deprecated":"Renamed to askedthrice-mcp (same tool, same maintainer). Install askedthrice-mcp instead; the endpoint is now https://mcp.askedthrice.com/mcp"}},"time":{"created":"2026-08-28T14:17:18.676Z","modified":"2026-08-30T18:47:31.207Z","0.1.0":"2026-08-28T14:17:19.016Z"},"bugs":{"url":"https://github.com/nahuelsoria/llmaudit-mcp/issues"},"author":{"name":"Nahuel Soria"},"license":"MIT","homepage":"https://llmaudit.app/developers","keywords":["mcp","model-context-protocol","mcp-server","ai-visibility","geo","generative-engine-optimization","brand-monitoring","llm","chatgpt","claude"],"repository":{"url":"git+https://github.com/nahuelsoria/llmaudit-mcp.git","type":"git","directory":"npm"},"description":"MCP server for llmaudit.app: measure whether AI assistants actually recommend a brand. stdio bridge to the hosted server, for clients that launch local commands.","maintainers":[{"name":"nahhuelsoria","email":"jorgenahuelsoria@gmail.com"}],"readme":"# llmaudit-mcp\n\nMCP server for [llmaudit.app](https://llmaudit.app). Ask your assistant whether a brand actually shows up when buyers ask AI for a recommendation, and get the measured answer: which buyer questions the brand won, which competitors were named instead, and which providers answered.\n\nFree, no signup, no API key. One measurement per domain every 30 days.\n\n## Install\n\nMost clients accept the hosted endpoint directly. Use that when you can:\n\n```json\n{\"mcpServers\":{\"llmaudit\":{\"url\":\"https://mcp.llmaudit.app/mcp\"}}}\n```\n\nClaude Code:\n\n```sh\nclaude mcp add --transport http llmaudit https://mcp.llmaudit.app/mcp\n```\n\nFor clients that only launch a local command (the Claude Desktop config file, for example), this package bridges stdio to the same server:\n\n```json\n{\"mcpServers\":{\"llmaudit\":{\"command\":\"npx\",\"args\":[\"-y\",\"llmaudit-mcp\"]}}}\n```\n\nRequires Node 18+. Nothing runs locally except the bridge: the measurement happens on the server, which holds the provider keys.\n\n## What it exposes\n\n- `start_visibility_check`: starts a measurement for a brand (name, website, category, optional competitors, location and language) and returns a `runId`. Takes about a minute, because the buyer questions are asked live to OpenAI, Anthropic and Gemini.\n- `get_visibility_check`: collects the result with the `runId`. Returns `running` until the providers answer, then a verdict in three bands and the counts.\n- `llmaudit://methodology`: how the measurement works.\n\nTry it: \"Measure how Acme (acme.com), a project management tool, shows up in AI recommendations against Asana and Monday.\"\n\n## What it returns, and what it does not\n\nIt returns the **measured map**: the buyer questions that were actually asked and who won each answer. It never returns a provider's self report (what a model *believes* about a brand), because in practice that number is the optimistic one and can be off by an order of magnitude. There is no 0 to 100 score either: a verdict in three bands and counts you can check.\n\nThere is no `email` field. The tool cannot send a result email to anyone.\n\n## Options\n\n- `LLMAUDIT_MCP_URL`: override the endpoint (staging).\n- Any extra argument is passed through to [mcp-remote](https://www.npmjs.com/package/mcp-remote), which does the transport.\n\n## Source\n\nServer code, tests and fixtures: [github.com/nahuelsoria/llmaudit-mcp](https://github.com/nahuelsoria/llmaudit-mcp). Developer docs: [llmaudit.app/developers](https://llmaudit.app/developers).\n","readmeFilename":"README.md"}