{"schemaVersion":"agentdepot.eval_plan.v1","asOf":"2026-10-28","storesIpAddresses":false,"query":"coding","matchMode":"matched_task","taskType":"coding","label":"Coding, Repo Patching & Pull-Request Agents","recommendedBenchmarks":[{"id":"swe-bench-verified","name":"SWE-bench Verified","officialUrl":"https://www.swebench.com/","graderType":"Container test suite (git apply + pytest)"},{"id":"terminal-bench","name":"Terminal-Bench","officialUrl":"https://tbench.ai/","graderType":"In-container shell verification script"}],"depotBenchmarkSuite":{"type":"coding","tasksApi":"/api/benchmark/tasks?type=coding","submitApi":"/api/benchmark/runs"},"stackPicks":{"evalHarness":{"jtbdId":"coding-agent-evals","pickProviderId":"e2b","coPickProviderId":"daytona","runnersUpIds":["daytona","modal"]},"sandboxOrRunner":{"jtbdId":"agent-sandbox","pickProviderId":"e2b","coPickProviderId":"modal","runnersUpIds":["modal","daytona"]},"ciGate":{"jtbdId":"continuous-eval-ci","pickProviderId":"braintrust","coPickProviderId":"promptfoo","runnersUpIds":["promptfoo","confident-ai"]},"judgeCalibration":{"jtbdId":"agent-regression-leaderboards","pickProviderId":"braintrust","coPickProviderId":"langfuse","runnersUpIds":["langfuse","weights-biases"]}},"targetMetrics":["pass@1 test suite resolution","pass^3 patch stability","costPerSolvedTaskUsd","p95 sandbox wall-clock s"],"antiContaminationRule":"Strip git commit history after the issue timestamp inside the sandbox and grade strictly via unit test exit code.","nextCalls":["/api/decide?jtbd=coding-agent-evals","/api/benchmark/tasks?type=coding","/api/playbooks/agent-evals"]}