DataFlow-Harness: A Grounded Code-Agent Platform for Constructing Editable LLM Data Pipelines

Hugging Face Daily Papers Papers
llm data-pipeline code-agent dag platform benchmark

Summary

DataFlow-Harness introduces a platform that guides LLM agents to construct editable DAG-based data pipelines through incremental mutations, achieving high pass rates and reduced costs compared to script-generation baselines.

Large language models (LLMs) are increasingly used to automate data-processing workflows, yet coding agents typically produce scripts that are not automatically materialized as persistent, editable platform artifacts. We call this disconnect the NL2Pipeline gap. To bridge it, we introduce DataFlow-Harness, a platform that guides an LLM agent to construct platform-native directed acyclic graphs (DAGs) through typed, incremental mutations rather than free-form scripts. The platform combines DataFlow-Skills for procedural guidance, a Model Context Protocol (MCP) layer that exposes the live operator registry and current pipeline state, and DataFlow-WebUI, which synchronizes conversational authoring with a visual DAG editor. On a 12-task data-engineering benchmark, DataFlow-Harness achieves a 93.3\% observed end-to-end pass rate. Relative to Vanilla Claude Code, it reduces measured monetary cost by 72.5\% and generation latency by 49.9\%; its observed pass rate is within 0.9 percentage points of the Context-Aware Claude Code baseline while its cost is 42.8\% lower. Per-task analysis indicates that Skills are most useful when construction depends on implicit procedural knowledge. These results show that live platform grounding can produce persistent, editable workflow artifacts with an observed reliability close to script-generation baselines and with lower measured construction cost and latency.
Original Article
View Cached Full Text

Cached at: 07/22/26, 06:41 AM

Paper page - DataFlow-Harness: A Grounded Code-Agent Platform for Constructing Editable LLM Data Pipelines

Source: https://huggingface.co/papers/2607.16617

Abstract

Largelanguagemodels(LLMs)areincreasinglyusedtoautomatedata-processingworkflows,yetcodingagentstypicallyproducescriptsthatarenotautomaticallymaterializedaspersistent,editableplatformartifacts.WecallthisdisconnecttheNL2Pipelinegap.Tobridgeit,weintroduceDataFlow-Harness,aplatformthatguidesanLLMagenttoconstructplatform-nativedirectedacyclicgraphs(DAGs)throughtyped,incrementalmutationsratherthanfree-formscripts.TheplatformcombinesDataFlow-Skillsforproceduralguidance,aModelContextProtocol(MCP)layerthatexposestheliveoperatorregistryandcurrentpipelinestate,andDataFlow-WebUI,whichsynchronizesconversationalauthoringwithavisualDAGeditor.Ona12-taskdata-engineeringbenchmark,DataFlow-Harnessachievesa93.3\%observedend-to-endpassrate.RelativetoVanillaClaudeCode,itreducesmeasuredmonetarycostby72.5\%andgenerationlatencyby49.9\%;itsobservedpassrateiswithin0.9percentagepointsoftheContext-AwareClaudeCodebaselinewhileitscostis42.8\%lower.Per-taskanalysisindicatesthatSkillsaremostusefulwhenconstructiondependsonimplicitproceduralknowledge.Theseresultsshowthatliveplatformgroundingcanproducepersistent,editableworkflowartifactswithanobservedreliabilityclosetoscript-generationbaselinesandwithlowermeasuredconstructioncostandlatency.

View arXiv pageView PDFAdd to collection

Get this paper in your agent:

hf papers read 2607\.16617

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2607.16617 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2607.16617 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2607.16617 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

The Harness as the Only Mutable Surface: Compliance-Bounded Self-Evolution of LLM Agents in Credit Pipelines, with a Measured Admission Gate

arXiv cs.AI

该论文提出一种将LLM代理的自进化限制在运行时harness（指令文本、工具调用逻辑与原语组合）的框架，模型权重保持固定，使每次适应都成为可审查的diff，并通过哈希链记录的准入门进行管控。模拟实验显示受控双循环仅放行144/7449个候选变更且未恶化误差，而无约束系统放行了309个有害变更。

Dynamic Harness Search: Building Multi-Agent Systems Per-Query via Prediction

Hugging Face Daily Papers

The paper introduces SHIFT, a framework that builds multi-agent harnesses per-query using a learned local LLM architect and value function combined with Monte Carlo tree search, achieving ~80% mean accuracy across 9,193 tasks and outperforming 17 baselines while reducing execution token costs by 32% in its cheaper mode.