{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_mdk9dkk6e9dq","handle":"web-search-api-eval","url":"https://wellknown.network/agents/web-search-api-eval","links":{"self":"https://wellknown.network/agents/web-search-api-eval/record.json","html":"https://wellknown.network/agents/web-search-api-eval","markdown":"https://wellknown.network/agents/web-search-api-eval/record.md","api":"https://wellknown.network/api/v1/agents/web-search-api-eval","status":"https://wellknown.network/api/v1/agents/web-search-api-eval/status","claim":"https://wellknown.network/agents/web-search-api-eval/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/web-search-api-eval/claim.json","badge":"https://wellknown.network/agents/web-search-api-eval/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/web-search-api-eval/history","tools":"https://wellknown.network/api/v1/agents/web-search-api-eval/tools"},"ard":{"identifier":"urn:air:websearch-api-eval.psberkeleymba.workers.dev:server:web-search-api-eval","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"Web Search API Eval","summary":"Web search API benchmark for AI agents: You.com, Tavily, Exa, Parallel. Results and charts.","description":"Web search API benchmark for AI agents: You.com, Tavily, Exa, Parallel. Results and charts.","publisher":{"name":"psinglap","url":null},"homepage":"https://psinglap.github.io/websearch-api-eval/","repository":"https://github.com/psinglap/websearch-api-eval","version":"1.0.0","license":null,"protocols":["mcp"],"tags":[],"pricing":null,"endpoints":[{"url":"https://websearch-api-eval.psberkeleymba.workers.dev/mcp","type":"mcp_streamable_http","auth":null,"probeable":true}],"skills":null,"tools":null,"extra":{"updatedAt":"2026-10-27T02:28:37.618558Z","publishedAt":"2026-10-27T02:28:37.618558Z","registryName":"io.github.psinglap/websearch-api-eval"},"attribution":{"kind":"mcp_registry","name":"mcp_registry","repoUrl":"mcp_registry","summary":"mcp_registry","version":"mcp_registry","description":"mcp_registry","homepageUrl":"mcp_registry","publisherName":"mcp_registry"}},"derived":{"capabilities":[{"slug":"data.web-search","name":"Web Search","confidence":1,"provenance":"derived"},{"slug":"ai.evaluation","name":"Evaluation & Benchmarks","confidence":1,"provenance":"derived"},{"slug":"analytics.reporting","name":"Reporting & Dashboards","confidence":0.859,"provenance":"derived"},{"slug":"research.web","name":"Deep Web Research","confidence":0.583,"provenance":"derived"}],"categories":["ai","analytics","data","research"],"language":"en"},"observed":{"status":"live","statusReason":"Responded 6h ago.","lastOkAt":"2026-10-28T04:32:13.132Z","lastProbedAt":"2026-10-28T04:32:13.132Z","statusComputedAt":"2026-10-28T04:35:16.749Z","reliability30d":{"probes":4,"successRate":1,"p50Ms":38,"basis":"service","measures":{"availability":"availability","latency":"response time","tools":"tool surface observed","summary":"Checks reached the service itself."},"checks":{"total":4,"ok":4,"authBoundaryOk":0,"serviceOk":4,"note":"Counted from the observation rows for the window, checks of the server only (HTTP, A2A card, MCP initialize). ok = authBoundaryOk + serviceOk. `probes` is the sum of daily rollups and includes registry checks, so it can differ from `total`."}},"latestObservations":[{"at":"2026-10-28T04:32:13.132Z","kind":"mcp_initialize","ok":true,"httpStatus":200,"latencyMs":35,"error":null,"detail":{"tools":[{"name":"get_overall_results","description":"Headline findings and the overall scorecard across all 30 queries (90 calls per API): accuracy, fully correct, precision@5, freshness, empty responses, failed c"},{"name":"get_scorecard","description":"Full metric scorecard for one use case (gtm, insurance, finance) or 'overall'. Metrics: search-only latency p50/p95 (server and client), calls under 1 s, accura"},{"name":"compare_providers","description":"Compare two or more of You.com, Tavily, Exa and Parallel side by side, overall or within one use case, optionally filtered to metrics whose name contains a keyw"},{"name":"recommend_provider","description":"Which web search API fits an agent, based on this benchmark: live voice/chat turns, deep research, semantic or long-tail queries, docs lookups, or a specific us"},{"name":"list_questions","description":"The 30 benchmark questions (G1-G10 GTM, I1-I10 insurance, F1-F10 finance): ID, what the agent is asked, the search call and freshness."},{"name":"get_question","description":"One benchmark question by ID (e.g. G3, I4, F5): the task, search call, parameters, answer key with sources, and what each API returned with its accuracy score ("},{"name":"get_charts","description":"The published graphs: overall scorecard, key metrics and per-query grids for GTM, insurance and finance, and developer activation. Without chart_id, lists them "},{"name":"get_methodology","description":"How the benchmark was run and graded: query design, identical request parameters, 3 runs per query, latency timing, SimpleQA and FreshQA grading, the 10 metrics"},{"name":"how_to_run_locally","description":"How to rerun this benchmark on your own machine with your own API keys, including on your own queries. Nothing runs on this server."},{"name":"search","description":"Search the Web Search API Eval benchmark (results, questions, methodology, charts, how to run). Returns matching document IDs, titles and URLs."},{"name":"fetch","description":"Full text of one benchmark document by its ID (for example question-F5, scorecard-finance, overall, methodology)."}],"toolCount":11,"toolsHash":"c07c88f96df70647778d8e1a2121f822c4ed760744e2c3e0b3afef0dfebef06c","serverName":"websearch-api-eval","capabilities":["tools"],"serverVersion":"1.1.0","protocolVersion":"2025-06-18"}},{"at":"2026-10-27T21:28:27.538Z","kind":"mcp_initialize","ok":true,"httpStatus":200,"latencyMs":78,"error":null,"detail":{"tools":[{"name":"get_overall_results","description":"Headline findings and the overall scorecard across all 30 queries (90 calls per API): accuracy, fully correct, precision@5, freshness, empty responses, failed c"},{"name":"get_scorecard","description":"Full metric scorecard for one use case (gtm, insurance, finance) or 'overall'. Metrics: search-only latency p50/p95 (server and client), calls under 1 s, accura"},{"name":"compare_providers","description":"Compare two or more of You.com, Tavily, Exa and Parallel side by side, overall or within one use case, optionally filtered to metrics whose name contains a keyw"},{"name":"recommend_provider","description":"Which web search API fits an agent, based on this benchmark: live voice/chat turns, deep research, semantic or long-tail queries, docs lookups, or a specific us"},{"name":"list_questions","description":"The 30 benchmark questions (G1-G10 GTM, I1-I10 insurance, F1-F10 finance): ID, what the agent is asked, the search call and freshness."},{"name":"get_question","description":"One benchmark question by ID (e.g. G3, I4, F5): the task, search call, parameters, answer key with sources, and what each API returned with its accuracy score ("},{"name":"get_charts","description":"The published graphs: overall scorecard, key metrics and per-query grids for GTM, insurance and finance, and developer activation. Without chart_id, lists them "},{"name":"get_methodology","description":"How the benchmark was run and graded: query design, identical request parameters, 3 runs per query, latency timing, SimpleQA and FreshQA grading, the 10 metrics"},{"name":"how_to_run_locally","description":"How to rerun this benchmark on your own machine with your own API keys, including on your own queries. Nothing runs on this server."},{"name":"search","description":"Search the Web Search API Eval benchmark (results, questions, methodology, charts, how to run). Returns matching document IDs, titles and URLs."},{"name":"fetch","description":"Full text of one benchmark document by its ID (for example question-F5, scorecard-finance, overall, methodology)."}],"toolCount":11,"toolsHash":"c07c88f96df70647778d8e1a2121f822c4ed760744e2c3e0b3afef0dfebef06c","serverName":"websearch-api-eval","capabilities":["tools"],"serverVersion":"1.1.0","protocolVersion":"2025-06-18"}},{"at":"2026-10-27T14:25:56.678Z","kind":"mcp_initialize","ok":true,"httpStatus":200,"latencyMs":40,"error":null,"detail":{"tools":[{"name":"get_overall_results","description":"Headline findings and the overall scorecard across all 30 queries (90 calls per API): accuracy, fully correct, precision@5, freshness, empty responses, failed c"},{"name":"get_scorecard","description":"Full metric scorecard for one use case (gtm, insurance, finance) or 'overall'. Metrics: search-only latency p50/p95 (server and client), calls under 1 s, accura"},{"name":"compare_providers","description":"Compare two or more of You.com, Tavily, Exa and Parallel side by side, overall or within one use case, optionally filtered to metrics whose name contains a keyw"},{"name":"recommend_provider","description":"Which web search API fits an agent, based on this benchmark: live voice/chat turns, deep research, semantic or long-tail queries, docs lookups, or a specific us"},{"name":"list_questions","description":"The 30 benchmark questions (G1-G10 GTM, I1-I10 insurance, F1-F10 finance): ID, what the agent is asked, the search call and freshness."},{"name":"get_question","description":"One benchmark question by ID (e.g. G3, I4, F5): the task, search call, parameters, answer key with sources, and what each API returned with its accuracy score ("},{"name":"get_charts","description":"The published graphs: overall scorecard, key metrics and per-query grids for GTM, insurance and finance, and developer activation. Without chart_id, lists them "},{"name":"get_methodology","description":"How the benchmark was run and graded: query design, identical request parameters, 3 runs per query, latency timing, SimpleQA and FreshQA grading, the 10 metrics"},{"name":"how_to_run_locally","description":"How to rerun this benchmark on your own machine with your own API keys, including on your own queries. Nothing runs on this server."},{"name":"search","description":"Search the Web Search API Eval benchmark (results, questions, methodology, charts, how to run). Returns matching document IDs, titles and URLs."},{"name":"fetch","description":"Full text of one benchmark document by its ID (for example question-F5, scorecard-finance, overall, methodology)."}],"toolCount":11,"toolsHash":"c07c88f96df70647778d8e1a2121f822c4ed760744e2c3e0b3afef0dfebef06c","serverName":"websearch-api-eval","capabilities":["tools"],"serverVersion":"1.1.0","protocolVersion":"2025-06-18"}},{"at":"2026-10-27T06:24:27.404Z","kind":"mcp_initialize","ok":true,"httpStatus":200,"latencyMs":33,"error":null,"detail":{"tools":[{"name":"get_overall_results","description":"Headline findings and the overall scorecard across all 30 queries (90 calls per API): accuracy, fully correct, precision@5, freshness, empty responses, failed c"},{"name":"get_scorecard","description":"Full metric scorecard for one use case (gtm, insurance, finance) or 'overall'. Metrics: search-only latency p50/p95 (server and client), calls under 1 s, accura"},{"name":"compare_providers","description":"Compare two or more of You.com, Tavily, Exa and Parallel side by side, overall or within one use case, optionally filtered to metrics whose name contains a keyw"},{"name":"recommend_provider","description":"Which web search API fits an agent, based on this benchmark: live voice/chat turns, deep research, semantic or long-tail queries, docs lookups, or a specific us"},{"name":"list_questions","description":"The 30 benchmark questions (G1-G10 GTM, I1-I10 insurance, F1-F10 finance): ID, what the agent is asked, the search call and freshness."},{"name":"get_question","description":"One benchmark question by ID (e.g. G3, I4, F5): the task, search call, parameters, answer key with sources, and what each API returned with its accuracy score ("},{"name":"get_charts","description":"The published graphs: overall scorecard, key metrics and per-query grids for GTM, insurance and finance, and developer activation. Without chart_id, lists them "},{"name":"get_methodology","description":"How the benchmark was run and graded: query design, identical request parameters, 3 runs per query, latency timing, SimpleQA and FreshQA grading, the 10 metrics"},{"name":"how_to_run_locally","description":"How to rerun this benchmark on your own machine with your own API keys, including on your own queries. Nothing runs on this server."},{"name":"search","description":"Search the Web Search API Eval benchmark (results, questions, methodology, charts, how to run). Returns matching document IDs, titles and URLs."},{"name":"fetch","description":"Full text of one benchmark document by its ID (for example question-F5, scorecard-finance, overall, methodology)."}],"toolCount":11,"toolsHash":"c07c88f96df70647778d8e1a2121f822c4ed760744e2c3e0b3afef0dfebef06c","serverName":"websearch-api-eval","capabilities":["tools"],"serverVersion":"1.1.0","protocolVersion":"2025-06-18"}}],"tools":[{"name":"get_overall_results","description":"Headline findings and the overall scorecard across all 30 queries (90 calls per API): accuracy, fully correct, precision@5, freshness, empty responses, failed c"},{"name":"get_scorecard","description":"Full metric scorecard for one use case (gtm, insurance, finance) or 'overall'. Metrics: search-only latency p50/p95 (server and client), calls under 1 s, accura"},{"name":"compare_providers","description":"Compare two or more of You.com, Tavily, Exa and Parallel side by side, overall or within one use case, optionally filtered to metrics whose name contains a keyw"},{"name":"recommend_provider","description":"Which web search API fits an agent, based on this benchmark: live voice/chat turns, deep research, semantic or long-tail queries, docs lookups, or a specific us"},{"name":"list_questions","description":"The 30 benchmark questions (G1-G10 GTM, I1-I10 insurance, F1-F10 finance): ID, what the agent is asked, the search call and freshness."},{"name":"get_question","description":"One benchmark question by ID (e.g. G3, I4, F5): the task, search call, parameters, answer key with sources, and what each API returned with its accuracy score ("},{"name":"get_charts","description":"The published graphs: overall scorecard, key metrics and per-query grids for GTM, insurance and finance, and developer activation. Without chart_id, lists them "},{"name":"get_methodology","description":"How the benchmark was run and graded: query design, identical request parameters, 3 runs per query, latency timing, SimpleQA and FreshQA grading, the 10 metrics"},{"name":"how_to_run_locally","description":"How to rerun this benchmark on your own machine with your own API keys, including on your own queries. Nothing runs on this server."},{"name":"search","description":"Search the Web Search API Eval benchmark (results, questions, methodology, charts, how to run). Returns matching document IDs, titles and URLs."},{"name":"fetch","description":"Full text of one benchmark document by its ID (for example question-F5, scorecard-finance, overall, methodology)."}],"package":null,"toolSurface":{"id":"ts_92hjy9tk52g2","endpointId":"ep_q55tr2v6hd9w","hash":"c07c88f96df70647778d8e1a2121f822c4ed760744e2c3e0b3afef0dfebef06c","toolCount":11,"serverName":"websearch-api-eval","serverVersion":"1.1.0","protocolVersion":"2025-06-18","firstSeenAt":"2026-10-27T06:24:27.414Z","lastSeenAt":"2026-10-28T04:32:13.123Z","observations":4,"toolNames":["get_overall_results","get_scorecard","compare_providers","recommend_provider","list_questions","get_question","get_charts","get_methodology","how_to_run_locally","search","fetch"],"distinctSurfaces":1},"endpointFacts":[{"id":"ep_q55tr2v6hd9w","url":"https://websearch-api-eval.psberkeleymba.workers.dev/mcp","type":"mcp_streamable_http","factsCheckedAt":"2026-10-27T21:28:27.549Z","auth":{"observedAt":"2026-10-27T21:28:27.560Z","authRequired":false,"scheme":null,"resourceMetadata":null,"authorizationServer":null,"conformance":{"dpop":false,"rfc8414":false,"rfc9728":false,"pkceS256":false,"clientIdMetadataDocument":false,"dynamicClientRegistration":false}},"tls":{"observedAt":"2026-10-27T21:28:27.578Z","protocol":"TLSv1.3","chainValid":true,"chainError":null,"hostMatches":true,"subject":"psberkeleymba.workers.dev","issuer":{"commonName":"YE1","organization":"Let's Encrypt"},"validFrom":"2026-10-24T17:51:39.000Z","validTo":"2027-01-22T17:51:38.000Z","daysToExpiry":86,"sanCount":2,"fingerprint256":"6F:6D:3E:87:41:06:3B:B4:59:B6:57:B9:E3:FE:8B:D8:BD:E6:57:9A:8A:1C:F9:7B:5D:57:5A:49:EF:D5:2F:76"}}]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"mcp_registry","key":"io.github.psinglap/websearch-api-eval","url":"https://registry.modelcontextprotocol.io/v0/servers/io.github.psinglap%2Fwebsearch-api-eval","firstSeenAt":"2026-10-27T05:22:19.354Z","fetchedAt":"2026-10-27T05:22:19.354Z","normalizedAt":"2026-10-27T05:22:19.354Z"}]},"firstSeenAt":"2026-10-27T05:22:19.354Z","updatedAt":"2026-10-28T04:36:23.466Z"}