{"schemaVersion":"agentdepot.eval_plan.v1","asOf":"2026-10-28","storesIpAddresses":false,"query":"rag","matchMode":"matched_task","taskType":"rag","label":"RAG, Deep Research, Grounded Synthesis & Citation Agents","recommendedBenchmarks":[{"id":"gaia","name":"GAIA (General AI Assistants)","officialUrl":"https://huggingface.co/spaces/gaia-benchmark/leaderboard","graderType":"Exact ground-truth answer + trajectory cost"}],"depotBenchmarkSuite":{"type":"research","tasksApi":"/api/benchmark/tasks?type=research","submitApi":"/api/benchmark/runs"},"stackPicks":{"evalHarness":{"jtbdId":"rag-evals","pickProviderId":"ragas","coPickProviderId":"promptfoo","runnersUpIds":["promptfoo","braintrust"]},"sandboxOrRunner":{"jtbdId":"public-agent-benchmarks","pickProviderId":"promptfoo","coPickProviderId":"braintrust","runnersUpIds":["braintrust","e2b"]},"ciGate":{"jtbdId":"llm-as-judge-calibration","pickProviderId":"braintrust","coPickProviderId":"patronus","runnersUpIds":["patronus","ragas"]},"judgeCalibration":{"jtbdId":"production-guardrail-evals","pickProviderId":"galileo","coPickProviderId":"braintrust","runnersUpIds":["braintrust","maxim"]}},"targetMetrics":["context precision & recall","citation faithfulness (0 fabricated URLs)","exact answer match on GAIA","costPerSolvedTaskUsd"],"antiContaminationRule":"Block benchmark leaderboard domains in search tools during eval runs and verify every cited claim against fetched source bytes.","nextCalls":["/api/decide?jtbd=rag-evals","/api/benchmark/tasks?type=research","/api/playbooks/agent-evals"]}