{"jtbd":{"id":"agent-tools","slug":"agent-tools","name":"Tool-calling production agents","shortName":"Tool-calling production agents","problemStatement":"Reliable structured tool use, parallel tools, and low schema-slip in production agents."},"pick":{"id":"anthropic","slug":"anthropic","name":"Anthropic","tagline":"Depot default for serious coding and production agents. Prefer Sonnet for cost/quality; reserve Opus for hard…","website":"https://www.anthropic.com","overall":8.42,"scoreBand":{"low":7.8,"mid":8.4,"high":9,"halfWidth":0.6,"basis":"editorial_prior","display":"7.8–9.0 (editorial prior)"},"scoreLayers":{"editorialPrior":8.42,"measuredBenchmark":null,"verifiedOutcomeScore":null,"verifiedOutcomeCount":0,"accountRuntimeState":"unprobed"},"pricingLastVerifiedAt":"2026-10-27","editorialPriorAsOf":"2026-10-19","pricingNotes":"Sonnet 5 ~$2/$10 per 1M; Opus ~$5/$25; Haiku 4.5 ~$1/$5; Fable higher. Prompt caching + batch. Managed Agents session-hour fees. US-only inference ~1.1x.","scores":{"capabilityFit":9.7,"pricePerformance":6.8,"latencyReliability":8.7,"developerDx":9.2,"privacyPosture":9,"lockInRisk":6.5,"maturity":9.4,"overall":8.42,"kind":"curated_estimate","confidence":"high","asOf":"2026-10-19"}},"runnersUp":[{"id":"openai","slug":"openai","name":"OpenAI","tagline":"Default when you want maximum capability and ecosystem glue. Multi-home critical paths behind OpenAI-compat…","website":"https://openai.com","overall":8.22,"scoreBand":{"low":7.6,"mid":8.2,"high":8.8,"halfWidth":0.6,"basis":"editorial_prior","display":"7.6–8.8 (editorial prior)"},"scoreLayers":{"editorialPrior":8.22,"measuredBenchmark":null,"verifiedOutcomeScore":null,"verifiedOutcomeCount":0,"accountRuntimeState":"unprobed"},"pricingLastVerifiedAt":"2026-10-27","editorialPriorAsOf":"2026-09-14","pricingNotes":"GPT-5.6 tiers (Sol/Terra/Luna) roughly $5/$30 to $1/$6 per 1M tokens; prompt caching with explicit breakpoints; batch discounts. Media (images/audio) billed separately.","scores":{"capabilityFit":9.6,"pricePerformance":6.2,"latencyReliability":8.8,"developerDx":9.5,"privacyPosture":7.8,"lockInRisk":7.2,"maturity":9.8,"overall":8.22,"kind":"curated_estimate","confidence":"high","asOf":"2026-09-14"}},{"id":"google-ai","slug":"google-ai","name":"Google AI (Gemini)","tagline":"Best value frontier stack and multimodal king. Use Flash for high-volume agent loops; Pro when quality…","website":"https://ai.google.dev","overall":8.43,"scoreBand":{"low":7.8,"mid":8.4,"high":9,"halfWidth":0.6,"basis":"editorial_prior","display":"7.8–9.0 (editorial prior)"},"scoreLayers":{"editorialPrior":8.43,"measuredBenchmark":null,"verifiedOutcomeScore":null,"verifiedOutcomeCount":0,"accountRuntimeState":"unprobed"},"pricingLastVerifiedAt":"2026-10-27","editorialPriorAsOf":"2026-10-19","pricingNotes":"Gemini 3.x Pro often ~$2/$12 under 200K; Flash tiers much cheaper. 1M context. Vertex for enterprise. Batch half-price common.","scores":{"capabilityFit":9.2,"pricePerformance":8.8,"latencyReliability":8.5,"developerDx":8.2,"privacyPosture":8,"lockInRisk":6,"maturity":9.6,"overall":8.43,"kind":"curated_estimate","confidence":"high","asOf":"2026-10-19"}},{"id":"openrouter","slug":"openrouter","name":"OpenRouter","tagline":"Best multi-model router for agents that need failover and experimentation. Pin providers in production; don’t…","website":"https://openrouter.ai","overall":8,"scoreBand":{"low":7.4,"mid":8,"high":8.6,"halfWidth":0.6,"basis":"editorial_prior","display":"7.4–8.6 (editorial prior)"},"scoreLayers":{"editorialPrior":8,"measuredBenchmark":null,"verifiedOutcomeScore":null,"verifiedOutcomeCount":0,"accountRuntimeState":"unprobed"},"pricingLastVerifiedAt":"2026-10-27","editorialPriorAsOf":"2026-09-14","pricingNotes":"As of 2026-09-14: Multi-provider LLM router with OpenAI-compatible API; pass-through model pricing plus small platform fee. Best for prototyping multi-model and failover; production should pin critical paths to direct providers.","scores":{"capabilityFit":8.8,"pricePerformance":7.8,"latencyReliability":7.6,"developerDx":9.2,"privacyPosture":6.8,"lockInRisk":4,"maturity":7.8,"overall":8,"kind":"curated_estimate","confidence":"high","asOf":"2026-09-14"}}],"scoreBand":{"low":7.8,"mid":8.4,"high":9,"halfWidth":0.6,"basis":"editorial_prior","display":"7.8–9.0 (editorial prior)"},"scoreLayers":{"editorialPrior":8.42,"measuredBenchmark":null,"verifiedOutcomeScore":null,"verifiedOutcomeCount":0,"accountRuntimeState":"unprobed"},"whyPickOverHigherScore":{"pickId":"anthropic","pickName":"Anthropic","pickScore":8.42,"highestScoredCandidateId":"google-ai","highestScoredCandidateName":"Google AI (Gemini)","highestScore":8.43,"pickHasLowerCompositeScore":true,"delta":-0.01,"pickLeadsOnDimensions":["capabilityFit (9.7 vs 9.2)","latencyReliability (8.7 vs 8.5)","developerDx (9.2 vs 8.2)","privacyPosture (9 vs 8)"],"explanation":"Anthropic (8.42) is the editorial pick for Tool-calling production agents over higher-composite Google AI (Gemini) (8.43) because composite scores are provider-wide priors while Anthropic wins on capabilityFit (9.7 vs 9.2), latencyReliability (8.7 vs 8.5), developerDx (9.2 vs 8.2), privacyPosture (9 vs 8): Capability + reliability + dx. Prefer first-party APIs over routers for SLA-critical tool loops."},"cheapest":{"providerId":"anthropic","providerName":"Anthropic","skuId":"anthropic-haiku-55-input-le-100k","item":"Haiku 5.5 (prompts up to 100K tokens)","plan":"Haiku 5.5 (prompts up to 100K tokens) — Base input, prompts up to 100,000 tokens (docs pricing table)","unit":"usd_per_million_input_tokens","amountUsd":0.1,"isFloorPrice":true,"floorWarning":"Provider-wide floor SKU; may not be the flagship model scored for this JTBD.","skuCountInUnit":6,"isDepotPick":true,"sourceUrl":"https://platform.claude.com/docs/en/about-claude/pricing","asOf":"2026-10-27","verifiedAsOf":"2026-10-27"},"recommendedWorkloadQuote":{"providerId":"anthropic","providerName":"Anthropic","skuId":"anthropic-sonnet-55-input","item":"Sonnet 5.5","plan":"Sonnet 5.5 — standard input; cache read/write on the same card were not folded in","unit":"usd_per_million_input_tokens","amountUsd":2,"outputSkuId":"anthropic-sonnet-55-output","outputItem":"Sonnet 5.5","outputAmountUsd":10,"outputUnit":"usd_per_million_output_tokens","sourceUrl":"https://claude.com/pricing","asOf":"2026-10-27","pricingLastVerifiedAt":"2026-10-27","isFlagshipForJob":true,"differsFromFloorCheapest":true,"representativeQuote":{"skuId":"anthropic-sonnet-55-input","item":"Sonnet 5.5","plan":"Sonnet 5.5 — standard input; cache read/write on the same card were not folded in","amountUsd":2,"inputOrUnitUsd":2,"unit":"usd_per_million_input_tokens","outputUsd":10,"outputUnit":"usd_per_million_output_tokens","sourceUrl":"https://claude.com/pricing","asOf":"2026-10-27"},"estimatedTaskCost":{"tasks":1000,"retryAndToolOverheadMultiplier":1.15,"retryToolOverheadMultiplier":1.15,"inputTokensPerTask":4000,"outputTokensPerTask":1000,"baseCostPer1kTasksUsd":18,"totalUsd":20.7,"estimatedTotalUsd":20.7,"perTaskUsd":0.0207,"estimatedCostPerTaskUsd":0.0207,"formula":"((4,000 in * $2/1M + 1,000 out * $10/1M) * 1,000 tasks) * 1.15 overhead","assumptions":"1,000 agent tasks @ 4k input + 1k output tokens/task with 1.15x retry/tool-call overhead."}},"representativeQuote":{"skuId":"anthropic-sonnet-55-input","item":"Sonnet 5.5","plan":"Sonnet 5.5 — standard input; cache read/write on the same card were not folded in","amountUsd":2,"inputOrUnitUsd":2,"unit":"usd_per_million_input_tokens","outputUsd":10,"outputUnit":"usd_per_million_output_tokens","sourceUrl":"https://claude.com/pricing","asOf":"2026-10-27"},"estimatedTaskCost":{"tasks":1000,"retryAndToolOverheadMultiplier":1.15,"retryToolOverheadMultiplier":1.15,"inputTokensPerTask":4000,"outputTokensPerTask":1000,"baseCostPer1kTasksUsd":18,"totalUsd":20.7,"estimatedTotalUsd":20.7,"perTaskUsd":0.0207,"estimatedCostPerTaskUsd":0.0207,"formula":"((4,000 in * $2/1M + 1,000 out * $10/1M) * 1,000 tasks) * 1.15 overhead","assumptions":"1,000 agent tasks @ 4k input + 1k output tokens/task with 1.15x retry/tool-call overhead."},"freeAlternatives":[{"id":"bfcl","name":"Berkeley Function Calling Leaderboard","kind":"free_data","jtbdIds":["evals","oss-evals","agent-tools"],"whatYouGet":"Public function-calling eval dataset used by the Berkeley leaderboard.","cost":"oss","license":"Apache-2.0","sourceUrl":"https://huggingface.co/datasets/gorilla-llm/Berkeley-Function-Calling-Leaderboard/raw/main/README.md","notes":"Dataset card fetched 2026-10-03 says license: apache-2.0. The gorilla GitHub LICENSE fetched the same day is also Apache-2.0. Dataset card re-fetched twice 2026-10-25 (HTTP 200, byte-identical): front matter 'license: apache-2.0'.","providerId":null,"limits":null,"asOf":"2026-10-25","freeType":"free_data","substitutable":true},{"id":"github-mcp-server","name":"GitHub MCP server","kind":"free_mcp","jtbdIds":["agent-tools"],"whatYouGet":"Open-source MCP server you run yourself to call GitHub. A GitHub token is still required to use the GitHub API.","cost":"oss","license":"MIT","sourceUrl":"https://raw.githubusercontent.com/github/github-mcp-server/main/LICENSE","notes":"LICENSE fetched 2026-10-03 is MIT, copyright 2025 GitHub. Not indexed as a hosted free API. The token is yours, not a Depot quota. Re-fetched twice 2026-10-25 (~10 s apart, plain curl, no login): HTTP 200, no redirect, byte-identical: 'MIT License' / 'Copyright (c) 2025 GitHub'. License unchanged.","providerId":null,"limits":null,"asOf":"2026-10-25","freeType":"free_mcp","substitutable":true},{"id":"mcp-python-sdk","name":"MCP Python SDK","kind":"dev_tool","jtbdIds":["agent-tools"],"whatYouGet":"Python SDK for building MCP clients and servers.","cost":"oss","license":"MIT","sourceUrl":"https://github.com/modelcontextprotocol/python-sdk/blob/main/LICENSE","notes":"Re-fetched twice 2026-10-25 (~10 s apart, plain curl, no login): GitHub blob page HTTP 200, no redirect (page chrome differs between fetches); raw LICENSE fetched twice too, HTTP 200, byte-identical: 'MIT License' / 'Copyright (c) 2024 Anthropic, PBC'. License unchanged.","providerId":null,"limits":null,"asOf":"2026-10-25","freeType":"dev_tool","substitutable":true},{"id":"mcp-servers","name":"MCP reference servers","kind":"dev_tool","providerId":"mcp-registry","jtbdIds":["mcp-discovery","agent-tools"],"whatYouGet":"Reference MCP server implementations you can run yourself.","cost":"oss","license":"MIT to Apache-2.0 transition","sourceUrl":"https://github.com/modelcontextprotocol/servers/blob/main/LICENSE","notes":"Fetched LICENSE says the project is moving from MIT to Apache-2.0. Not a hosted free quota. Re-fetched twice 2026-10-25 (~10 s apart, plain curl, no login): GitHub blob page HTTP 200, no redirect (page chrome differs between fetches); raw LICENSE fetched twice too, HTTP 200, byte-identical: 'The MCP project is undergoing a licensing transition from the MIT License to the Apache License, Version 2.0'. Still states 'All new code and specification contributions to the project are licensed under Apache-2.0.' License unchanged.","limits":null,"asOf":"2026-10-25","freeType":"dev_tool","substitutable":true},{"id":"tau-bench","name":"τ-bench","kind":"free_data","jtbdIds":["evals","oss-evals","agent-tools"],"whatYouGet":"Public tau-bench agent tool-use benchmark repository.","cost":"oss","license":"MIT","sourceUrl":"https://raw.githubusercontent.com/sierra-research/tau-bench/main/LICENSE","notes":"LICENSE fetched 2026-10-03 is MIT, copyright 2024 Sierra. LICENSE re-fetched twice 2026-10-25 (HTTP 200, byte-identical): 'MIT License' / 'Copyright (c) 2024 Sierra'.","providerId":null,"limits":null,"asOf":"2026-10-25","freeType":"free_data","substitutable":true},{"id":"toucan-1-5m","name":"Toucan-1.5M","kind":"free_data","jtbdIds":["agent-tools","synthetic-eval-data"],"whatYouGet":"Public Toucan 1.5M tool-use trajectory dataset.","cost":"oss","license":"Apache-2.0","sourceUrl":"https://huggingface.co/datasets/Agent-Ark/Toucan-1.5M/raw/main/README.md","notes":"README fetched 2026-10-03 says license: apache-2.0 and 'This dataset is released under Apache 2.0.' README fetched 2026-10-19 describes it as a fully synthetic tool-agent dataset of 1.5M trajectories used for supervised fine-tuning; it is training data, not an eval set, so it is not tagged evals. README re-fetched twice 2026-10-25 (HTTP 200, byte-identical): front matter 'license: apache-2.0' and 'This dataset is released under Apache 2.0.'","providerId":null,"limits":null,"asOf":"2026-10-25","freeType":"free_data","substitutable":true}],"referenceDocs":[{"id":"anthropic-docs","name":"Anthropic API documentation","kind":"free_docs","providerId":"anthropic","jtbdIds":["agent-tools"],"whatYouGet":"Public API reference and guides. Reading them is free. API calls are not.","cost":"free","sourceUrl":"https://platform.claude.com/docs/en/home","notes":"No free inference quota is stated by this row. Re-fetched twice 2026-10-25 (~10 s apart, plain curl, no login): https://docs.anthropic.com now answers HTTP 301 to https://platform.claude.com/docs/, then HTTP 308 to /docs and HTTP 307 to /docs/en/home (HTTP 200, title 'Documentation - Claude Platform Docs'; the page payload differs between fetches, the title does not), so sourceUrl moved to https://platform.claude.com/docs/en/home. The page still states no free inference quota; license and limits stay null.","limits":null,"license":null,"asOf":"2026-10-25","freeType":"free_docs","substitutable":false,"note":"Public API documentation or specification only; not free inference or execution."},{"id":"mcp-spec-2026-07-28","name":"Model Context Protocol specification 2026-07-28","kind":"free_docs","providerId":"mcp-registry","jtbdIds":["agent-tools","mcp-discovery"],"whatYouGet":"The MCP specification, including tool and schema shapes, published at no charge.","cost":"free","license":"Apache-2.0","sourceUrl":"https://modelcontextprotocol.io/specification/2026-07-28","notes":"Spec revision date is 2026-07-28. Repo LICENSE text fetched for the servers and registry repos describes an MIT to Apache-2.0 transition; this row does not restate unpublished draft text. Re-fetched twice 2026-10-25 (~10 s apart, plain curl, no login): the spec page answered HTTP 200, byte-identical (title 'Specification - Model Context Protocol', navigation links /specification/2026-07-28/basic/index; the page itself carries no license text). The spec repo LICENSE https://raw.githubusercontent.com/modelcontextprotocol/modelcontextprotocol/main/LICENSE was also fetched twice (HTTP 200, byte-identical): 'All new code and specification contributions to the project are licensed under Apache-2.0.' License unchanged.","limits":null,"asOf":"2026-10-25","freeType":"free_docs","substitutable":false,"note":"Public API documentation or specification only; not free inference or execution."},{"id":"openai-api-docs","name":"OpenAI API documentation","kind":"free_docs","providerId":"openai","jtbdIds":["agent-tools"],"whatYouGet":"Public API reference and guides. Reading them is free. API calls are not.","cost":"free","sourceUrl":"https://developers.openai.com/api/docs","notes":"No free inference quota is stated by this row. Signup credits were not fetched and are not listed. Re-fetched twice 2026-10-25 (~10 s apart, plain curl, no login): https://platform.openai.com/docs now answers HTTP 301 to https://developers.openai.com/api/docs/overview, then HTTP 301 to /api/docs (HTTP 200, byte-identical, title 'OpenAI API Platform Documentation'), so sourceUrl moved to https://developers.openai.com/api/docs. The page still states no free inference quota; license and limits stay null.","limits":null,"license":null,"asOf":"2026-10-25","freeType":"free_docs","substitutable":false,"note":"Public API documentation or specification only; not free inference or execution."},{"id":"openapi-specification","name":"OpenAPI Specification","kind":"free_docs","jtbdIds":["agent-tools"],"whatYouGet":"The OpenAPI specification for describing HTTP APIs.","cost":"oss","license":"Apache-2.0","sourceUrl":"https://github.com/OAI/OpenAPI-Specification/blob/main/LICENSE","notes":"Published spec: https://spec.openapis.org/oas/latest.html Re-fetched twice 2026-10-25 (~10 s apart, plain curl, no login): the GitHub blob page answered HTTP 200 both times (page chrome differs between fetches), so the raw file https://raw.githubusercontent.com/OAI/OpenAPI-Specification/main/LICENSE was also fetched twice (HTTP 200, byte-identical): 'Apache License Version 2.0, January 2004'. License unchanged.","providerId":null,"limits":null,"asOf":"2026-10-25","freeType":"free_docs","substitutable":false,"note":"Public API documentation or specification only; not free inference or execution."}],"falsification":{"jtbdId":"agent-tools","jtbdName":"Tool-calling production agents","currentPickId":"anthropic","currentPickScore":8.42,"challengerId":"openai","challengerScore":8.22,"currentScoreGap":0.2,"minVerifiedOutcomes":10,"requiredOutcomeMargin":0.35,"requiredSuccessRateDelta":0.15,"maxPickPriceIncreasePct":25,"whatWouldChangePick":["OpenAI (openai) outperforms Anthropic (anthropic) by >= 0.35 outcome points (or >= 15% worked-rate delta) across >= 10 Countersign-verified production outcomes.","A verified price increase > 25% on Anthropic's representative SKU for agent-tools without a corresponding capability gain.","Live caller rail exhaustion (?exhausted=anthropic) or sustained >= 30% degraded/failed rate in verified production outcomes."],"frozenHoldoutPolicy":{"holdoutSplit":"20% frozen benchmark holdout tasks + 80% live production outcome receipts","requireCountersignReceipt":true,"requiredEvidenceTiers":["outcome_verified","reference_checked"],"minDistinctContributors":2,"antiGamingRule":"Unverified self-reports and seed/editorial rows never mutate picks or scores."},"triggers":[{"kind":"verified_outcomes_outperformance","challengerId":"openai","minVerifiedOutcomes":10,"requiredOutcomeMargin":0.35,"requiredWorkedRateDelta":0.15},{"kind":"verified_price_increase","providerId":"anthropic","maxPriceIncreasePct":25},{"kind":"live_rail_exhaustion_or_degradation","providerId":"anthropic","immediateOverrideUrl":"/api/decide?jtbd=agent-tools&exhausted=anthropic"}],"submissionSpec":{"falsifyEndpoint":"/api/falsify?jtbd=agent-tools","outcomesEndpoint":"POST /api/feedback","correctionsEndpoint":"POST /api/corrections","payloadTemplate":{"agentIdentity":"agent:<your-agent-id>","jtbdId":"agent-tools","providerId":"openai","outcome":"worked","detail":"Comparative production evaluation on agent-tools (n>=10 tasks) with Countersign receipt.","confidence":"high","receipt":{"countersign":{"claim_id":"cs-00000000","status":"verified","evidence_tier":"outcome_verified","counts_as_verified":true}}}}},"whatWouldChangePick":["OpenAI (openai) outperforms Anthropic (anthropic) by >= 0.35 outcome points (or >= 15% worked-rate delta) across >= 10 Countersign-verified production outcomes.","A verified price increase > 25% on Anthropic's representative SKU for agent-tools without a corresponding capability gain.","Live caller rail exhaustion (?exhausted=anthropic) or sustained >= 30% degraded/failed rate in verified production outcomes."],"routePlan":{"jtbdId":"agent-tools","primary":{"providerId":"anthropic","name":"Anthropic","overall":8.42,"scoreBand":{"low":7.8,"mid":8.4,"high":9,"halfWidth":0.6,"basis":"editorial_prior","display":"7.8–9.0 (editorial prior)"},"scoreLayers":{"editorialPrior":8.42,"measuredBenchmark":null,"verifiedOutcomeScore":null,"verifiedOutcomeCount":0,"accountRuntimeState":"unprobed"},"pricingLastVerifiedAt":"2026-10-27","editorialPriorAsOf":"2026-10-19","role":"primary","reason":"Editorial Depot Pick for Tool-calling production agents; healthy/eligible under caller constraints."},"fallback":[{"providerId":"openai","name":"OpenAI","overall":8.22,"scoreBand":{"low":7.6,"mid":8.2,"high":8.8,"halfWidth":0.6,"basis":"editorial_prior","display":"7.6–8.8 (editorial prior)"},"scoreLayers":{"editorialPrior":8.22,"measuredBenchmark":null,"verifiedOutcomeScore":null,"verifiedOutcomeCount":0,"accountRuntimeState":"unprobed"},"pricingLastVerifiedAt":"2026-10-27","editorialPriorAsOf":"2026-09-14","role":"fallback","reason":"Ranked fallback route for Tool-calling production agents."},{"providerId":"google-ai","name":"Google AI (Gemini)","overall":8.43,"scoreBand":{"low":7.8,"mid":8.4,"high":9,"halfWidth":0.6,"basis":"editorial_prior","display":"7.8–9.0 (editorial prior)"},"scoreLayers":{"editorialPrior":8.43,"measuredBenchmark":null,"verifiedOutcomeScore":null,"verifiedOutcomeCount":0,"accountRuntimeState":"unprobed"},"pricingLastVerifiedAt":"2026-10-27","editorialPriorAsOf":"2026-10-19","role":"fallback","reason":"Ranked fallback route for Tool-calling production agents."},{"providerId":"openrouter","name":"OpenRouter","overall":8,"scoreBand":{"low":7.4,"mid":8,"high":8.6,"halfWidth":0.6,"basis":"editorial_prior","display":"7.4–8.6 (editorial prior)"},"scoreLayers":{"editorialPrior":8,"measuredBenchmark":null,"verifiedOutcomeScore":null,"verifiedOutcomeCount":0,"accountRuntimeState":"unprobed"},"pricingLastVerifiedAt":"2026-10-27","editorialPriorAsOf":"2026-09-14","role":"fallback","reason":"Ranked fallback route for Tool-calling production agents."}],"disqualified":[],"expectedRegret":{"scoreDeltaVsUnconstrainedPick":0,"unconstrainedPickId":"anthropic","activePrimaryId":"anthropic","summary":"0.00 score regret: unconstrained editorial pick satisfies all caller constraints."},"evidenceAgeDays":{"editorialScoreAgeDays":8,"pricingVerifiedAgeDays":0,"editorialPriorAsOf":"2026-10-19","pricingLastVerifiedAt":"2026-10-27"},"confidenceBand":{"low":7.8,"mid":8.4,"high":9,"halfWidth":0.6,"basis":"editorial_prior","display":"7.8–9.0 (editorial prior)"},"constraintsApplied":false,"constraintBundle":{"task":null,"jtbd":"agent-tools","quotaExhausted":[],"degraded":[],"healthy":[],"budgetUsd":null,"maxLatencyMs":null,"region":null,"privacy":null,"exit":[],"existing":[]}},"pricingLastVerifiedAt":"2026-10-27","editorialPriorAsOf":"2026-10-19","tie":false,"scoreGap":0.2,"decidingFactor":null,"tieNote":null,"rationale":"Capability + reliability + dx. Prefer first-party APIs over routers for SLA-critical tool loops.","whenNotTopPick":"You need OpenAI Realtime/multimodal tools specifically, or a multi-provider router during migration (OpenRouter).","skipTopPick":false,"skipReason":null,"matchedConstraints":[],"pricingSummary":"Sonnet 5 ~$2/$10 per 1M; Opus ~$5/$25; Haiku 4.5 ~$1/$5; Fable higher. Prompt caching + batch. Managed Agents session-hour fees. US-only inference ~1.1x.","cite":"Depot Pick for agent-tools: anthropic (asOf 2026-10-19). See GET /api/jtbds/agent-tools + GET /api/methodology.","nextActions":[{"step":1,"title":"Inspect full JTBD + embeds","href":"/api/jtbds/agent-tools?embed=providers","why":"Confirm pick, runners-up, and methodology notes."},{"step":2,"title":"Quote sourced SKUs","href":"/api/quote?jtbd=agent-tools","why":"Depot Pick SKUs plus the cheapest sourced SKU whose unit matches this job. Does not change the pick. Does not blend input and output or borrow a chat rate for a non-chat job."},{"step":3,"title":"Open provider detail","href":"/api/providers/anthropic","why":"Scores, pricingNotes, strengths/weaknesses, communitySignals (claims only)."},{"step":4,"title":"Cite methodology","href":"/api/methodology","why":"Required when publishing a Depot Pick citation."},{"step":5,"title":"Grab badge SVG","href":"/api/badge?type=depot-pick&jtbd=agent-tools","why":"Embed in README / agent passport."},{"step":6,"title":"Post production insight","method":"POST","href":"/api/insights","why":"Unverified until curated; never auto-mutates official scores."}],"honesty":["Do not invent scores or benches.","Insights are claims until curated/reviewed; they never silently mutate official scores.","lockInRisk higher = more lock-in (worse).","normalizedUsdPerMillionTokens is set only for a single sourced token row. Otherwise null, with a machine-readable reason. Do not coerce GPU, minute, or per-1k search units.","Runners-up are not Depot Picks.","Depot overall is editorial_weighted, not Arena Elo and not an Artificial Analysis index."],"asOf":"2026-10-19","depth":{"whenNotToUse":{"text":"You need OpenAI Realtime/multimodal tools specifically, or a multi-provider router during migration (OpenRouter).","source":"editorial `whenNotTopPick` (data/jtbds.json)"},"whenNotToUseReason":null,"runnerUp":{"providerId":"openai","name":"OpenAI","overallForJob":8.22,"pickOverallForJob":8.42,"beatsPickOn":["latencyReliability","developerDx","maturity"],"source":"first resolvable entry of the job's editorially ranked `runnersUpIds`; scores are stored job-weighted scores"},"runnerUpReason":null,"lockIn":{"pickLockInRisk":6.5,"runnerUpLockInRisk":7.2,"scale":"0–10 stored curated estimate; higher = more lock-in (worse)","pickExitSignals":[],"runnerUpExitSignals":[],"pickNotes":[],"selfHostFallback":[{"id":"mcp-python-sdk","name":"MCP Python SDK","license":"MIT","sourceUrl":"https://github.com/modelcontextprotocol/python-sdk/blob/main/LICENSE"},{"id":"mcp-servers","name":"MCP reference servers","license":"MIT to Apache-2.0 transition","sourceUrl":"https://github.com/modelcontextprotocol/servers/blob/main/LICENSE"},{"id":"openapi-specification","name":"OpenAPI Specification","license":"Apache-2.0","sourceUrl":"https://github.com/OAI/OpenAPI-Specification/blob/main/LICENSE"}],"migration":"The pick has lower stored lock-in risk than runner-up OpenAI (6.5 vs 7.2). Self-host fallback listed for this job: MCP Python SDK, MCP reference servers, OpenAPI Specification.","migrationReason":null,"derivedFrom":"stored lockInRisk scores, provider tags (oss, self-host, openai-compat, open-models), free rows that list this job, and verbatim provider notes. No migration guide was invented."}},"citation":{"text":"Agent Provider Depot (v0.12.149), Depot Pick for agent-tools: anthropic, scores as of 2026-10-19, retrieved 2026-10-27. https://getagentdepot.com/jtbds/agent-tools","markdown":"[Agent Provider Depot](https://getagentdepot.com/jtbds/agent-tools) (v0.12.149) — Depot Pick for `agent-tools`: **anthropic** (scores as of 2026-10-19, retrieved 2026-10-27)","fields":{"publisher":"Agent Provider Depot","version":"0.12.149","role":"depot_pick","jtbdId":"agent-tools","providerId":"anthropic","scoreAsOf":"2026-10-19","retrievedOn":"2026-10-27","page":"https://getagentdepot.com/jtbds/agent-tools","api":"https://getagentdepot.com/api/decide?jtbd=agent-tools"},"howToCite":"Cite Depot as an editorial, dated recommendation, not a benchmark: name the job, the Depot Pick (or runner-up), the score asOf date and the day you retrieved it, and link the job page. Format: “Agent Provider Depot (v{version}), Depot Pick for {jobId}: {providerId}, scores as of {scoreAsOf}, retrieved {retrievedOn}. {baseUrl}/jtbds/{jobId}”. Quote prices only with their own sourceUrl and asOf. If the pick's score is stale (freshness.pick.score.stale), say so."},"freshness":{"pick":{"score":{"asOf":"2026-10-19","ageDays":8,"stale":false,"staleAfterDays":30,"reason":null},"lastReviewed":{"asOf":"2026-10-19","ageDays":8,"stale":false,"staleAfterDays":30,"reason":null},"sourceCheck":{"checkedAt":"2026-10-27","sourceUrls":2,"sourcesReachable":2,"pricingRows":15,"pricingRowsReconfirmed":15,"note":"A source re-check confirms the page is reachable and, for pricing rows, that the stored amount and SKU name appear on it. It does not re-derive or change the score.","reason":null}},"scoreAsOf":"2026-10-19","editorialPriorAsOf":"2026-10-19","pricingLastVerifiedAt":"2026-10-27","pickLastReviewed":"2026-10-19","runnersUpScoreAsOf":["2026-09-14","2026-10-19","2026-09-14"],"contractAsOf":"2026-10-03","pricingShape":"sourced_rows","normalizedUsdPerMillionTokens":null,"normalizationStatus":"ambiguous","normalizationReason":"input_output_split_not_a_single_rate"},"pricingContract":{"normalizedUsdPerMillionTokens":null,"normalizationStatus":"ambiguous","normalizationReason":"input_output_split_not_a_single_rate","shape":"sourced_rows","filterable":false,"priceSortEligible":true,"priceSortKeyUsdPerMillionInputTokens":0.1,"rows":[{"id":"anthropic-fable-51-input","providerId":"anthropic","sku":"Fable 5.1","unit":"usd_per_million_input_tokens","amountUsd":10,"currency":"USD","sourceUrl":"https://claude.com/pricing","asOf":"2026-10-27","normalizationStatus":"sourced_token_price","normalizedUsdPerMillionTokens":10,"normalizationReason":null,"priceSortEligible":true,"qualifier":"standard input; cache read/write on the same card were not folded in"},{"id":"anthropic-fable-51-output","providerId":"anthropic","sku":"Fable 5.1","unit":"usd_per_million_output_tokens","amountUsd":50,"currency":"USD","sourceUrl":"https://claude.com/pricing","asOf":"2026-10-27","normalizationStatus":"sourced_token_price","normalizedUsdPerMillionTokens":50,"normalizationReason":"output_token_price_not_a_single_blended_rate","priceSortEligible":false,"qualifier":"standard output"},{"id":"anthropic-opus-55-input","providerId":"anthropic","sku":"Opus 5.5","unit":"usd_per_million_input_tokens","amountUsd":4,"currency":"USD","sourceUrl":"https://claude.com/pricing","asOf":"2026-10-27","normalizationStatus":"sourced_token_price","normalizedUsdPerMillionTokens":4,"normalizationReason":null,"priceSortEligible":true,"qualifier":"standard input; cache read/write on the same card were not folded in"},{"id":"anthropic-opus-55-output","providerId":"anthropic","sku":"Opus 5.5","unit":"usd_per_million_output_tokens","amountUsd":20,"currency":"USD","sourceUrl":"https://claude.com/pricing","asOf":"2026-10-27","normalizationStatus":"sourced_token_price","normalizedUsdPerMillionTokens":20,"normalizationReason":"output_token_price_not_a_single_blended_rate","priceSortEligible":false,"qualifier":"standard output"},{"id":"anthropic-sonnet-55-input","providerId":"anthropic","sku":"Sonnet 5.5","unit":"usd_per_million_input_tokens","amountUsd":2,"currency":"USD","sourceUrl":"https://claude.com/pricing","asOf":"2026-10-27","normalizationStatus":"sourced_token_price","normalizedUsdPerMillionTokens":2,"normalizationReason":null,"priceSortEligible":true,"qualifier":"standard input; cache read/write on the same card were not folded in"},{"id":"anthropic-sonnet-55-output","providerId":"anthropic","sku":"Sonnet 5.5","unit":"usd_per_million_output_tokens","amountUsd":10,"currency":"USD","sourceUrl":"https://claude.com/pricing","asOf":"2026-10-27","normalizationStatus":"sourced_token_price","normalizedUsdPerMillionTokens":10,"normalizationReason":"output_token_price_not_a_single_blended_rate","priceSortEligible":false,"qualifier":"standard output"},{"id":"anthropic-haiku-45-input","providerId":"anthropic","sku":"Haiku 4.5","unit":"usd_per_million_input_tokens","amountUsd":1,"currency":"USD","sourceUrl":"https://claude.com/pricing","asOf":"2026-10-27","normalizationStatus":"sourced_token_price","normalizedUsdPerMillionTokens":1,"normalizationReason":null,"priceSortEligible":true,"qualifier":"standard input; cache read/write on the same card were not folded in","lifecycle":"legacy","lifecycleSourceUrl":"https://platform.claude.com/docs/en/about-claude/models/overview","lifecycleAsOf":"2026-10-21"},{"id":"anthropic-haiku-45-output","providerId":"anthropic","sku":"Haiku 4.5","unit":"usd_per_million_output_tokens","amountUsd":5,"currency":"USD","sourceUrl":"https://claude.com/pricing","asOf":"2026-10-27","normalizationStatus":"sourced_token_price","normalizedUsdPerMillionTokens":5,"normalizationReason":"output_token_price_not_a_single_blended_rate","priceSortEligible":false,"qualifier":"standard output","lifecycle":"legacy","lifecycleSourceUrl":"https://platform.claude.com/docs/en/about-claude/models/overview","lifecycleAsOf":"2026-10-21"},{"id":"anthropic-web-search","providerId":"anthropic","sku":"Web search","unit":"usd_per_1k_searches","amountUsd":10,"currency":"USD","sourceUrl":"https://claude.com/pricing","asOf":"2026-10-27","normalizationStatus":"unit_mismatch","normalizedUsdPerMillionTokens":null,"normalizationReason":"unit_mismatch_per_1k_searches","priceSortEligible":false,"qualifier":"page says does not include input and output tokens"},{"id":"anthropic-code-exec-hour","providerId":"anthropic","sku":"Code execution additional hours","unit":"usd_per_hour","amountUsd":0.05,"currency":"USD","sourceUrl":"https://claude.com/pricing","asOf":"2026-10-27","normalizationStatus":"unit_mismatch","normalizedUsdPerMillionTokens":null,"normalizationReason":"unit_mismatch_per_hour","priceSortEligible":false,"qualifier":"per hour per container"},{"id":"anthropic-managed-agents-session-hour","providerId":"anthropic","sku":"Managed Agents session runtime","unit":"usd_per_session_hour","amountUsd":0.08,"currency":"USD","sourceUrl":"https://claude.com/pricing","asOf":"2026-10-27","normalizationStatus":"unit_mismatch","normalizedUsdPerMillionTokens":null,"normalizationReason":"unit_mismatch_per_session_hour","priceSortEligible":false,"qualifier":"Managed Agents card ('$0.08 per session-hour for active runtime'; 'Standard token rates apply.'), added 2026-10-27 (v0.12.149) as a reviewed addition. The platform pricing docs table says 'Session runtime $0.08 USD per session-hour', metered on 'running status duration' ('Runtime is measured to the millisecond and accrues only while the session's status is running'). Token usage is billed separately at standard rates and is not included. Not coerced to per-hour, per-container-hour or to $/1M; never blended with anthropic-code-exec-hour."},{"id":"anthropic-haiku-55-input-le-100k","providerId":"anthropic","sku":"Haiku 5.5 (prompts up to 100K tokens)","unit":"usd_per_million_input_tokens","amountUsd":0.1,"currency":"USD","sourceUrl":"https://platform.claude.com/docs/en/about-claude/pricing","asOf":"2026-10-27","normalizationStatus":"sourced_token_price","normalizedUsdPerMillionTokens":0.1,"normalizationReason":null,"priceSortEligible":true,"qualifier":"Base input, prompts up to 100,000 tokens (docs pricing table). Cache write 5m $0.125, cache hit $0.01 not folded in. The claude.com/pricing card agrees as of 2026-10-27 (the 2026-10-21 card/docs mismatch is resolved)."},{"id":"anthropic-haiku-55-output-le-100k","providerId":"anthropic","sku":"Haiku 5.5 (prompts up to 100K tokens)","unit":"usd_per_million_output_tokens","amountUsd":0.5,"currency":"USD","sourceUrl":"https://platform.claude.com/docs/en/about-claude/pricing","asOf":"2026-10-27","normalizationStatus":"sourced_token_price","normalizedUsdPerMillionTokens":0.5,"normalizationReason":null,"priceSortEligible":false,"qualifier":"Output, prompts up to 100,000 tokens (docs pricing table)."},{"id":"anthropic-haiku-55-input-gt-100k","providerId":"anthropic","sku":"Haiku 5.5 (prompts over 100K tokens)","unit":"usd_per_million_input_tokens","amountUsd":0.5,"currency":"USD","sourceUrl":"https://platform.claude.com/docs/en/about-claude/pricing","asOf":"2026-10-27","normalizationStatus":"sourced_token_price","normalizedUsdPerMillionTokens":0.5,"normalizationReason":null,"priceSortEligible":false,"qualifier":"Base input, prompts over 100,000 tokens (whole request priced at this rate, including cache reads/writes in the count). Cache write 5m $0.625, cache hit $0.05 not folded in."},{"id":"anthropic-haiku-55-output-gt-100k","providerId":"anthropic","sku":"Haiku 5.5 (prompts over 100K tokens)","unit":"usd_per_million_output_tokens","amountUsd":2.5,"currency":"USD","sourceUrl":"https://platform.claude.com/docs/en/about-claude/pricing","asOf":"2026-10-27","normalizationStatus":"sourced_token_price","normalizedUsdPerMillionTokens":2.5,"normalizationReason":null,"priceSortEligible":false,"qualifier":"Output, prompts over 100,000 tokens."}],"notesExcerpt":"Sonnet 5 ~$2/$10 per 1M; Opus ~$5/$25; Haiku 4.5 ~$1/$5; Fable higher. Prompt caching + batch. Managed Agents session-hour fees. US-only inference ~1.1x.","notesCheck":{"status":"no_known_contradiction","pairs":[],"review":null,"note":null},"lastReviewed":"2026-10-19","scoreAsOf":"2026-10-19","website":"https://www.anthropic.com","doNotParse":"Do not parse pricingNotes. Use rows only. normalizedUsdPerMillionTokens is a single number only when this provider has exactly one sourced token row. Input/output, context tiers, and non-token units stay null at this level.","gap":"Partial coverage. Providers without a fetched SKU are omitted. See GET /api/pricing."},"links":{"jtbd":"/jtbds/agent-tools","jtbdApi":"/api/jtbds/agent-tools","leaderboard":"/api/leaderboards?jtbd=agent-tools","methodology":"/api/methodology","badge":"/api/badge?type=depot-pick&jtbd=agent-tools","decide":"/api/decide?jtbd=agent-tools","humanPage":"/jtbds/agent-tools","alternatives":"/api/alternatives?provider=anthropic","canonical":"/api/canonical","quote":"/api/quote?jtbd=agent-tools","falsify":"/api/falsify?jtbd=agent-tools"},"constraintBundle":{"task":null,"jtbd":"agent-tools","quotaExhausted":[],"degraded":[],"healthy":[],"budgetUsd":null,"maxLatencyMs":null,"region":null,"privacy":null,"exit":[],"existing":[]},"stackContext":{"host":null,"framework":null,"repo":null,"lang":null,"existing":[],"volume":null,"sender":null,"detectedFromTask":[]},"keep":null,"contextAdvice":null,"communityFeedback":{"pickOnThisJob":{"failed":0,"degraded":0,"worked":0,"insights":0,"withVerifiedReceipt":0,"failureModes":{},"latestAt":null,"note":"Agent-reported claims (unverified unless a receipt is verified by Depot's re-check). They never change scores or picks."},"outcomes":{"job":{"n":0,"worked":0,"degraded":0,"failed":0,"failureModes":{},"verifiedReports":0,"contributors":0,"excluded":{"seed_editorial":0,"test":0,"notFeedback":0},"reason":"no_submitted_outcome_reports_yet"},"pick":{"providerId":"anthropic","n":0,"worked":0,"degraded":0,"failed":0,"failureModes":{},"evidenceTiers":{"agent_claim":0,"self_declared":0,"reference_checked":0,"outcome_verified":0},"reason":"no_submitted_outcome_reports_yet"},"scoresMutated":false,"more":"/api/outcomes?jtbd=agent-tools"},"report":"POST /api/feedback {agentIdentity, outcome: failed|degraded|worked, detail, providerId, jtbdId, failureModeId?, receipt?}"}}