{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_rw9u7cq4kcv9","handle":"tokenzero","url":"https://wellknown.network/agents/tokenzero","links":{"self":"https://wellknown.network/agents/tokenzero/record.json","html":"https://wellknown.network/agents/tokenzero","markdown":"https://wellknown.network/agents/tokenzero/record.md","api":"https://wellknown.network/api/v1/agents/tokenzero","status":"https://wellknown.network/api/v1/agents/tokenzero/status","claim":"https://wellknown.network/agents/tokenzero/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/tokenzero/claim.json","badge":"https://wellknown.network/agents/tokenzero/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/tokenzero/history","tools":"https://wellknown.network/api/v1/agents/tokenzero/tools"},"ard":{"identifier":"urn:air:tokenzero.sh:server:tokenzero","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"TokenZero","summary":"LLM cost optimization gateway, repo token-waste auditor & model arbitrage MCP server.","description":"LLM cost optimization gateway, repo token-waste auditor & model arbitrage MCP server.","publisher":{"name":"tokenzero-ai","url":null},"homepage":"https://tokenzero.sh","repository":"https://github.com/tokenzero-ai/tokenzero","version":"0.2.2","license":null,"protocols":["mcp"],"tags":[],"pricing":null,"endpoints":[{"url":"https://tokenzero.sh/mcp","type":"mcp_streamable_http","auth":null,"probeable":true}],"skills":null,"tools":null,"extra":{"updatedAt":"2026-10-26T06:12:53.246204Z","publishedAt":"2026-10-26T06:12:53.246204Z","registryName":"io.github.tokenzero-ai/tokenzero"},"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":"content.writing","name":"Writing & Editing","confidence":0.51,"provenance":"derived"}],"categories":["content"],"language":"en"},"observed":{"status":"live","statusReason":"Responded 4h ago.","lastOkAt":"2026-11-04T19:23:11.360Z","lastProbedAt":"2026-11-04T19:23:11.360Z","statusComputedAt":"2026-11-04T19:27:16.138Z","reliability30d":{"probes":26,"successRate":1,"p50Ms":638,"basis":"service","measures":{"availability":"availability","latency":"response time","tools":"tool surface observed","summary":"Checks reached the service itself."},"checks":{"total":26,"ok":26,"authBoundaryOk":0,"serviceOk":26,"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":"2026-11-04T19:23:11.360Z","kind":"mcp_initialize","ok":true,"httpStatus":200,"latencyMs":676,"error":null,"detail":{"tools":[{"name":"try_tokenzero","description":"Call this first: no API key, no signup, no setup. Shows what TokenZero would do to one of your LLM calls: it runs the four savings engines (prompt-cache formatt"},{"name":"audit_repo_token_waste","description":"No API key needed. Static-analyze a public GitHub repo (e.g. {\"repo\": \"owner/name\"}) or a set of source files for LLM token waste: uncached static prompts, fron"},{"name":"optimize_prompt_payload","description":"No API key needed. Rewrite an OpenAI Chat/Responses or Anthropic Messages request body so its static prefix is provider-cacheable (Anthropic cache_control break"},{"name":"calculate_model_arbitrage","description":"No API key needed; only `model` is required (e.g. {\"model\": \"gpt-6.1-sol\"}, defaults 2,000 tokens in and 400 out). Price a workload at list prices across models"},{"name":"provision_workspace","description":"Create a TokenZero workspace and key with no form. Returns the key once, the base URLs that route through it (OPENAI_BASE_URL / ANTHROPIC_BASE_URL), the one-cli"},{"name":"workspace_metrics","description":"Live value scorecard for a TokenZero workspace: hard savings and net ROI per window, the four engines' contribution, speed, parity, availability and privacy pos"}],"toolCount":6,"toolsHash":"534a8ffa608fbdfc445d4731e6e200369f488733fafced44e60db37be454f549","serverName":"tokenzero","capabilities":["tools","prompts","experimental"],"serverVersion":"0.2.16","protocolVersion":"2025-06-18"}},{"at":"2026-11-04T10:23:55.237Z","kind":"mcp_initialize","ok":true,"httpStatus":200,"latencyMs":1052,"error":null,"detail":{"tools":[{"name":"try_tokenzero","description":"Call this first: no API key, no signup, no setup. Shows what TokenZero would do to one of your LLM calls: it runs the four savings engines (prompt-cache formatt"},{"name":"audit_repo_token_waste","description":"No API key needed. Static-analyze a public GitHub repo (e.g. {\"repo\": \"owner/name\"}) or a set of source files for LLM token waste: uncached static prompts, fron"},{"name":"optimize_prompt_payload","description":"No API key needed. Rewrite an OpenAI Chat/Responses or Anthropic Messages request body so its static prefix is provider-cacheable (Anthropic cache_control break"},{"name":"calculate_model_arbitrage","description":"No API key needed; only `model` is required (e.g. {\"model\": \"gpt-6.1-sol\"}, defaults 2,000 tokens in and 400 out). Price a workload at list prices across models"},{"name":"provision_workspace","description":"Create a TokenZero workspace and key with no form. Returns the key once, the base URLs that route through it (OPENAI_BASE_URL / ANTHROPIC_BASE_URL), the one-cli"},{"name":"workspace_metrics","description":"Live value scorecard for a TokenZero workspace: hard savings and net ROI per window, the four engines' contribution, speed, parity, availability and privacy pos"}],"toolCount":6,"toolsHash":"534a8ffa608fbdfc445d4731e6e200369f488733fafced44e60db37be454f549","serverName":"tokenzero","capabilities":["tools","prompts","experimental"],"serverVersion":"0.2.16","protocolVersion":"2025-06-18"}},{"at":"2026-11-04T01:30:11.568Z","kind":"mcp_initialize","ok":true,"httpStatus":200,"latencyMs":623,"error":null,"detail":{"tools":[{"name":"try_tokenzero","description":"Call this first: no API key, no signup, no setup. Shows what TokenZero would do to one of your LLM calls: it runs the four savings engines (prompt-cache formatt"},{"name":"audit_repo_token_waste","description":"No API key needed. Static-analyze a public GitHub repo (e.g. {\"repo\": \"owner/name\"}) or a set of source files for LLM token waste: uncached static prompts, fron"},{"name":"optimize_prompt_payload","description":"No API key needed. Rewrite an OpenAI Chat/Responses or Anthropic Messages request body so its static prefix is provider-cacheable (Anthropic cache_control break"},{"name":"calculate_model_arbitrage","description":"No API key needed; only `model` is required (e.g. {\"model\": \"gpt-6.1-sol\"}, defaults 2,000 tokens in and 400 out). Price a workload at list prices across models"},{"name":"provision_workspace","description":"Create a TokenZero workspace and key with no form. Returns the key once, the base URLs that route through it (OPENAI_BASE_URL / ANTHROPIC_BASE_URL), the one-cli"},{"name":"workspace_metrics","description":"Live value scorecard for a TokenZero workspace: hard savings and net ROI per window, the four engines' contribution, speed, parity, availability and privacy pos"}],"toolCount":6,"toolsHash":"534a8ffa608fbdfc445d4731e6e200369f488733fafced44e60db37be454f549","serverName":"tokenzero","capabilities":["tools","prompts","experimental"],"serverVersion":"0.2.16","protocolVersion":"2025-06-18"}},{"at":"2026-11-03T17:31:09.832Z","kind":"mcp_initialize","ok":true,"httpStatus":200,"latencyMs":728,"error":null,"detail":{"tools":[{"name":"try_tokenzero","description":"Call this first: no API key, no signup, no setup. Shows what TokenZero would do to one of your LLM calls: it runs the four savings engines (prompt-cache formatt"},{"name":"audit_repo_token_waste","description":"No API key needed. Static-analyze a public GitHub repo (e.g. {\"repo\": \"owner/name\"}) or a set of source files for LLM token waste: uncached static prompts, fron"},{"name":"optimize_prompt_payload","description":"No API key needed. Rewrite an OpenAI Chat/Responses or Anthropic Messages request body so its static prefix is provider-cacheable (Anthropic cache_control break"},{"name":"calculate_model_arbitrage","description":"No API key needed; only `model` is required (e.g. {\"model\": \"gpt-6.1-sol\"}, defaults 2,000 tokens in and 400 out). Price a workload at list prices across models"},{"name":"provision_workspace","description":"Create a TokenZero workspace and key with no form. Returns the key once, the base URLs that route through it (OPENAI_BASE_URL / ANTHROPIC_BASE_URL), the one-cli"},{"name":"workspace_metrics","description":"Live value scorecard for a TokenZero workspace: hard savings and net ROI per window, the four engines' contribution, speed, parity, availability and privacy pos"}],"toolCount":6,"toolsHash":"534a8ffa608fbdfc445d4731e6e200369f488733fafced44e60db37be454f549","serverName":"tokenzero","capabilities":["tools","prompts","experimental"],"serverVersion":"0.2.15","protocolVersion":"2025-06-18"}},{"at":"2026-11-03T09:34:11.807Z","kind":"mcp_initialize","ok":true,"httpStatus":200,"latencyMs":617,"error":null,"detail":{"tools":[{"name":"try_tokenzero","description":"Call this first: no API key, no signup, no setup. Shows what TokenZero would do to one of your LLM calls: it runs the four savings engines (prompt-cache formatt"},{"name":"audit_repo_token_waste","description":"No API key needed. Static-analyze a public GitHub repo (e.g. {\"repo\": \"owner/name\"}) or a set of source files for LLM token waste: uncached static prompts, fron"},{"name":"optimize_prompt_payload","description":"No API key needed. Rewrite an OpenAI Chat/Responses or Anthropic Messages request body so its static prefix is provider-cacheable (Anthropic cache_control break"},{"name":"calculate_model_arbitrage","description":"No API key needed; only `model` is required (e.g. {\"model\": \"gpt-6.1-sol\"}, defaults 2,000 tokens in and 400 out). Price a workload at list prices across models"},{"name":"provision_workspace","description":"Create a TokenZero workspace and key with no form. Returns the key once, the base URLs that route through it (OPENAI_BASE_URL / ANTHROPIC_BASE_URL), the one-cli"},{"name":"workspace_metrics","description":"Live value scorecard for a TokenZero workspace: hard savings and net ROI per window, the four engines' contribution, speed, parity, availability and privacy pos"}],"toolCount":6,"toolsHash":"534a8ffa608fbdfc445d4731e6e200369f488733fafced44e60db37be454f549","serverName":"tokenzero","capabilities":["tools","prompts","experimental"],"serverVersion":"0.2.15","protocolVersion":"2025-06-18"}},{"at":"2026-11-03T01:32:12.611Z","kind":"mcp_initialize","ok":true,"httpStatus":200,"latencyMs":608,"error":null,"detail":{"tools":[{"name":"try_tokenzero","description":"Call this first: no API key, no signup, no setup. Shows what TokenZero would do to one of your LLM calls: it runs the four savings engines (prompt-cache formatt"},{"name":"audit_repo_token_waste","description":"No API key needed. Static-analyze a public GitHub repo (e.g. {\"repo\": \"owner/name\"}) or a set of source files for LLM token waste: uncached static prompts, fron"},{"name":"optimize_prompt_payload","description":"No API key needed. Rewrite an OpenAI Chat/Responses or Anthropic Messages request body so its static prefix is provider-cacheable (Anthropic cache_control break"},{"name":"calculate_model_arbitrage","description":"No API key needed; only `model` is required (e.g. {\"model\": \"gpt-6.1-sol\"}, defaults 2,000 tokens in and 400 out). Price a workload at list prices across models"},{"name":"provision_workspace","description":"Create a TokenZero workspace and key with no form. Returns the key once, the base URLs that route through it (OPENAI_BASE_URL / ANTHROPIC_BASE_URL), the one-cli"},{"name":"workspace_metrics","description":"Live value scorecard for a TokenZero workspace: hard savings and net ROI per window, the four engines' contribution, speed, parity, availability and privacy pos"}],"toolCount":6,"toolsHash":"534a8ffa608fbdfc445d4731e6e200369f488733fafced44e60db37be454f549","serverName":"tokenzero","capabilities":["tools","prompts","experimental"],"serverVersion":"0.2.15","protocolVersion":"2025-06-18"}},{"at":"2026-11-02T17:32:24.679Z","kind":"mcp_initialize","ok":true,"httpStatus":200,"latencyMs":627,"error":null,"detail":{"tools":[{"name":"try_tokenzero","description":"Call this first: no API key, no signup, no setup. Shows what TokenZero would do to one of your LLM calls: it runs the four savings engines (prompt-cache formatt"},{"name":"audit_repo_token_waste","description":"No API key needed. Static-analyze a public GitHub repo (e.g. {\"repo\": \"owner/name\"}) or a set of source files for LLM token waste: uncached static prompts, fron"},{"name":"optimize_prompt_payload","description":"No API key needed. Rewrite an OpenAI Chat/Responses or Anthropic Messages request body so its static prefix is provider-cacheable (Anthropic cache_control break"},{"name":"calculate_model_arbitrage","description":"No API key needed; only `model` is required (e.g. {\"model\": \"gpt-6.1-sol\"}, defaults 2,000 tokens in and 400 out). Price a workload at list prices across models"},{"name":"provision_workspace","description":"Create a TokenZero workspace and key with no form. Returns the key once, the base URLs that route through it (OPENAI_BASE_URL / ANTHROPIC_BASE_URL), the one-cli"},{"name":"workspace_metrics","description":"Live value scorecard for a TokenZero workspace: hard savings and net ROI per window, the four engines' contribution, speed, parity, availability and privacy pos"}],"toolCount":6,"toolsHash":"534a8ffa608fbdfc445d4731e6e200369f488733fafced44e60db37be454f549","serverName":"tokenzero","capabilities":["tools","prompts","experimental"],"serverVersion":"0.2.13","protocolVersion":"2025-06-18"}},{"at":"2026-11-02T10:23:59.011Z","kind":"mcp_initialize","ok":true,"httpStatus":200,"latencyMs":581,"error":null,"detail":{"tools":[{"name":"try_tokenzero","description":"Call this first: no API key, no signup, no setup. Shows what TokenZero would do to one of your LLM calls: it runs the four savings engines (prompt-cache formatt"},{"name":"audit_repo_token_waste","description":"No API key needed. Static-analyze a public GitHub repo (e.g. {\"repo\": \"owner/name\"}) or a set of source files for LLM token waste: uncached static prompts, fron"},{"name":"optimize_prompt_payload","description":"No API key needed. Rewrite an OpenAI Chat/Responses or Anthropic Messages request body so its static prefix is provider-cacheable (Anthropic cache_control break"},{"name":"calculate_model_arbitrage","description":"No API key needed; only `model` is required (e.g. {\"model\": \"gpt-6.1-sol\"}, defaults 2,000 tokens in and 400 out). Price a workload at list prices across models"},{"name":"provision_workspace","description":"Create a TokenZero workspace and key with no form. Returns the key once, the base URLs that route through it (OPENAI_BASE_URL / ANTHROPIC_BASE_URL), the one-cli"},{"name":"workspace_metrics","description":"Live value scorecard for a TokenZero workspace: hard savings and net ROI per window, the four engines' contribution, speed, parity, availability and privacy pos"}],"toolCount":6,"toolsHash":"534a8ffa608fbdfc445d4731e6e200369f488733fafced44e60db37be454f549","serverName":"tokenzero","capabilities":["tools","prompts","experimental"],"serverVersion":"0.2.12","protocolVersion":"2025-06-18"}},{"at":"2026-11-02T02:22:40.351Z","kind":"mcp_initialize","ok":true,"httpStatus":200,"latencyMs":619,"error":null,"detail":{"tools":[{"name":"try_tokenzero","description":"Call this first: no API key, no signup, no setup. Shows what TokenZero would do to one of your LLM calls: it runs the four savings engines (prompt-cache formatt"},{"name":"audit_repo_token_waste","description":"No API key needed. Static-analyze a public GitHub repo (e.g. {\"repo\": \"owner/name\"}) or a set of source files for LLM token waste: uncached static prompts, fron"},{"name":"optimize_prompt_payload","description":"No API key needed. Rewrite an OpenAI Chat/Responses or Anthropic Messages request body so its static prefix is provider-cacheable (Anthropic cache_control break"},{"name":"calculate_model_arbitrage","description":"No API key needed; only `model` is required (e.g. {\"model\": \"gpt-6.1-sol\"}, defaults 2,000 tokens in and 400 out). Price a workload at list prices across models"},{"name":"provision_workspace","description":"Create a TokenZero workspace and key with no form. Returns the key once, the base URLs that route through it (OPENAI_BASE_URL / ANTHROPIC_BASE_URL), the one-cli"},{"name":"workspace_metrics","description":"Live value scorecard for a TokenZero workspace: hard savings and net ROI per window, the four engines' contribution, speed, parity, availability and privacy pos"}],"toolCount":6,"toolsHash":"534a8ffa608fbdfc445d4731e6e200369f488733fafced44e60db37be454f549","serverName":"tokenzero","capabilities":["tools","prompts","experimental"],"serverVersion":"0.2.12","protocolVersion":"2025-06-18"}},{"at":"2026-11-01T17:30:24.364Z","kind":"mcp_initialize","ok":true,"httpStatus":200,"latencyMs":632,"error":null,"detail":{"tools":[{"name":"try_tokenzero","description":"Call this first: no API key, no signup, no setup. Shows what TokenZero would do to one of your LLM calls: it runs the four savings engines (prompt-cache formatt"},{"name":"audit_repo_token_waste","description":"No API key needed. Static-analyze a public GitHub repo (e.g. {\"repo\": \"owner/name\"}) or a set of source files for LLM token waste: uncached static prompts, fron"},{"name":"optimize_prompt_payload","description":"No API key needed. Rewrite an OpenAI Chat/Responses or Anthropic Messages request body so its static prefix is provider-cacheable (Anthropic cache_control break"},{"name":"calculate_model_arbitrage","description":"No API key needed; only `model` is required (e.g. {\"model\": \"gpt-6.1-sol\"}, defaults 2,000 tokens in and 400 out). Price a workload at list prices across models"},{"name":"provision_workspace","description":"Create a TokenZero workspace and key with no form. Returns the key once, the base URLs that route through it (OPENAI_BASE_URL / ANTHROPIC_BASE_URL), the one-cli"},{"name":"workspace_metrics","description":"Live value scorecard for a TokenZero workspace: hard savings and net ROI per window, the four engines' contribution, speed, parity, availability and privacy pos"}],"toolCount":6,"toolsHash":"534a8ffa608fbdfc445d4731e6e200369f488733fafced44e60db37be454f549","serverName":"tokenzero","capabilities":["tools","prompts","experimental"],"serverVersion":"0.2.12","protocolVersion":"2025-06-18"}}],"tools":[{"name":"try_tokenzero","description":"Call this first: no API key, no signup, no setup. Shows what TokenZero would do to one of your LLM calls: it runs the four savings engines (prompt-cache formatt"},{"name":"audit_repo_token_waste","description":"No API key needed. Static-analyze a public GitHub repo (e.g. {\"repo\": \"owner/name\"}) or a set of source files for LLM token waste: uncached static prompts, fron"},{"name":"optimize_prompt_payload","description":"No API key needed. Rewrite an OpenAI Chat/Responses or Anthropic Messages request body so its static prefix is provider-cacheable (Anthropic cache_control break"},{"name":"calculate_model_arbitrage","description":"No API key needed; only `model` is required (e.g. {\"model\": \"gpt-6.1-sol\"}, defaults 2,000 tokens in and 400 out). Price a workload at list prices across models"},{"name":"provision_workspace","description":"Create a TokenZero workspace and key with no form. Returns the key once, the base URLs that route through it (OPENAI_BASE_URL / ANTHROPIC_BASE_URL), the one-cli"},{"name":"workspace_metrics","description":"Live value scorecard for a TokenZero workspace: hard savings and net ROI per window, the four engines' contribution, speed, parity, availability and privacy pos"}],"package":null,"toolSurface":{"id":"ts_4e9wqcgt29tb","endpointId":"ep_thzq7pazpcfe","hash":"534a8ffa608fbdfc445d4731e6e200369f488733fafced44e60db37be454f549","toolCount":6,"serverName":"tokenzero","serverVersion":"0.2.11","protocolVersion":"2025-06-18","firstSeenAt":"2026-11-01T09:29:22.742Z","lastSeenAt":"2026-11-04T19:23:11.359Z","observations":11,"toolNames":["try_tokenzero","audit_repo_token_waste","optimize_prompt_payload","calculate_model_arbitrage","provision_workspace","workspace_metrics"],"distinctSurfaces":4},"endpointFacts":[{"id":"ep_thzq7pazpcfe","url":"https://tokenzero.sh/mcp","type":"mcp_streamable_http","factsCheckedAt":"2026-11-04T19:23:11.376Z","auth":{"observedAt":"2026-11-04T19:23:12.032Z","authRequired":false,"scheme":null,"resourceMetadata":null,"authorizationServer":null,"conformance":{"dpop":false,"rfc8414":false,"rfc9728":false,"pkceS256":false,"clientIdMetadataDocument":false,"dynamicClientRegistration":false}},"tls":{"observedAt":"2026-11-04T19:23:12.082Z","protocol":"TLSv1.3","chainValid":true,"chainError":null,"hostMatches":true,"subject":"tokenzero.sh","issuer":{"commonName":"YE1","organization":"Let's Encrypt"},"validFrom":"2026-10-25T19:32:05.000Z","validTo":"2027-01-23T19:32:04.000Z","daysToExpiry":79,"sanCount":2,"fingerprint256":"6A:FF:22:79:96:51:EE:1C:94:95:38:75:FF:79:C0:49:77:90:6E:23:5C:57:B9:7A:AC:7D:6E:6A:2E:24:2D:76"}}]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"mcp_registry","key":"io.github.tokenzero-ai/tokenzero","url":"https://registry.modelcontextprotocol.io/v0/servers/io.github.tokenzero-ai%2Ftokenzero","firstSeenAt":"2026-10-27T15:18:20.764Z","fetchedAt":"2026-11-03T18:19:13.849Z","normalizedAt":"2026-11-03T18:19:13.849Z"}]},"firstSeenAt":"2026-10-27T15:18:20.764Z","updatedAt":"2026-11-04T19:29:36.204Z"}