AI Daybook
AI Daybook
Today's issue
PaperAgents

HEELO uses agent-native primitives to automate real-world tasks

Original titleThe Last Mile to the Real World: Agent-Native Primitives as Reusable Building Blocks for Solving Real-World Tasks

Outlet
Hugging Face
Published
Importance
2/5
Trending
12

Summary

The authors propose agent-native primitives, reusable units equipped with a resident LLM that applies resources in context and communicates with other primitives. They introduce HEELO, a framework that composes these primitives into task-specific graphs and revises them based on environmental feedback. On Terminal-Bench 4.0, the approach raises resolution from 37.3% to 55.5%, and on 46 real-world tasks, Claude Opus 5.5 with Claude Code and GPT-6.1 Sol with Codex complete 39 and 35 tasks respectively when connected to HEELO, compared to 9 each on their own.

Original excerpt

How can an agent handle a real-world task end to end? Frontier models with their own harnesses still fall short on real-world benchmarks: on Terminal-Bench 4.0, GPT-5.6 Sol with Codex resolves 37.3% of the tasks and Claude Opus 5 with Claude Code 51.8%. These tasks fail at the step from reaching a resource to applying it, which we call the last mile to the real world. Existing approaches put the resource within reach but leave the competence to apply it with the caller, so each harness rebuilds that competence. We introduce agent-native primitives, reusable units that equip a resource with a resident LLM that applies it in context and communicates with other primitives, and HEELO (Harness Engineering around Evolving Agent-Native Primitives for the Last Mile to Real-WOrld Tasks), a harness engineering framework that builds each task's harness from a library of such primitives. HEELO composes the primitives a task needs into a task-specific graph, verifies the result against evidence from the environment, revises the primitives a failure implicates, and admits new or improved primitives after validation. We instantiate HEELO with eight primitive families, each built for one kind of task, and evaluate them on more than twenty benchmarks. Dev-Primitives raise Terminal-Bench 4.0 resolution from 37.3% under Codex to 55.5% on the same backbone, Harness Primitives composed per task raise Terminal-Bench 2.0 Pass@1 from 60.0% for the seed harness to 72.2%, Tool Primitives raise ToolBench pass rate from 55.8% to 75.1%, and Navigation Primitives raise GAIA-Text-103 accuracy from 45.6% to 58.3%. On the 46 real-world tasks of our heelo.dev platform, Claude Opus 5.5 with Claude Code and GPT-6.1 Sol with Codex complete 9 each on their own and 39 and 35 when connected to HEELO through our MCP server, with 1.9times and 2.3times less time and 1.55times lower cost per completed task.

#agent#benchmark#lmm
Read the original
See all →