{ "schema_version": 1, "builder_plan": { "scenario": "Your coding assistant makes changes that still need substantial repair before merging.", "change": "Try the technique on held-out issues from a repository with reliable tests.", "baseline": "The same coding agent, repository snapshot, tools, and token budget.", "steps": [ "Locate the paper's implementation and check its license, model support, data needs, and hardware requirements. Stop here if you cannot obtain the required artifacts.", "Save a small set of representative product tasks and expected outcomes. Keep evaluation tasks out of any training or tuning, and choose quality and cost limits before running.", "Run the baseline and the proposed change on identical inputs. Keep other settings fixed and repeat paired runs to expose variation.", "Save per-task outputs, failures, timings, and all usage costs. Review regressions before expanding the test." ], "measure": "Correct patches, regression failures, human repair time, and total cost per accepted change.", "decision": "Keep the change only if it meets your preselected quality and latency limits and improves the metric you care about after setup and operating costs. Otherwise retain the baseline. This is an exploration plan, not a validated recommendation." }, "builder_review": null, "arxiv_id": "2609.20519", "title": "SoL-Pi: Recursively Scaling Auto-Research Loops for Efficient Agent Harness", "summary": "SoL-Pi replaces the default agent harness (action execution, context compaction, observation handling, and delegated reading in tools such as Codex or Claude Code) with four mechanisms selected by recursive auto-research rollouts to cut token traffic at equal task performance. The cost is integration effort at the harness layer and dependence on mechanisms discovered through automated selection, plus added complexity in how observations and context are compacted for the model. Failure modes include compaction or delegated reading dropping information needed on task distributions unlike EdgeBench, mechanisms that may not transfer to other model providers or harnesses, and selection bias toward the 51-task evaluation used to validate them.", "rationale": "The harness-level changes are compatible with API-only constrained infrastructure and claim quality-neutral token savings, but validation rests on a single 51-task benchmark, so the reader should pilot it on their own coding workloads before production adoption.", "technique": "Token-efficient agent harness via four mechanisms", "family": "agentes_de_codigo", "recommendation": "testar", "published": "2026-09-17", "as_of": "2026-10-18", "editorial_status": "indexed", "source_url": "https://arxiv.org/abs/2609.20519", "independent_implementations": 9, "total_implementations": 10, "stars": 9065, "citations": 0, "claims": [ { "claim_id": "claim-01", "statement": "SoL-Pi replaces the default agent harness (action execution, context compaction, observation handling, and delegated reading in tools such as Codex or Claude Code) with four mechanisms selected by recursive auto-research rollouts to cut token traffic at equal task performance. The cost is integration effort at the harness layer and dependence on mechanisms discovered through automated selection, plus added complexity in how observations and context are compacted for the model. Failure modes include compaction or delegated reading dropping information needed on task distributions unlike EdgeBench, mechanisms that may not transfer to other model providers or harnesses, and selection bias toward the 51-task evaluation used to validate them.", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2609.20519", "source_page": null, "source_excerpt": "" }, { "claim_id": "claim-02", "statement": "On the 51-task EdgeBench evaluation, SoL-Pi matches Pi-level performance across GPT-5.6 Sol and Opus 5 while reducing recorded token traffic by 44.7-49.0% and API cost by about one third (estimated $8.75-$13.50/hour vs. native Codex and Claude Code harnesses, $4.36-$5.71/hour vs. Pi).", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2609.20519", "source_page": null, "source_excerpt": "" } ], "exposure_map": [ { "dimension": "quality", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "compute", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "latency", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "operations", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "compatibility", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "security", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "data_and_training", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" }, { "dimension": "reproducibility", "basis": "not_evaluated", "finding": "", "interpretation": "", "limitation": "", "next_check": "", "source_url": "", "source_page": null, "source_excerpt": "" } ], "risks": [], "minimum_test": [], "open_questions": [], "independent_tests": [], "lab_reproductions": [], "report_available": false, "validation_tier": "unknown", "evidence_tier": "unknown", "infrastructure_basis": "unknown", "equations": [], "equations_status": "not_fetched", "core_kind": "", "equations_fetched_at": "", "equations_source_url": "", "equations_source_sha256": "", "exposures_reviewed_at": "", "exposures_source_sha256": "" }