Harness Handbook uses behavior-centric repo maps to help coding agents edit faster
rohanpaul_ai · x · 2026-07-24
- This paper argues that agent harnesses are hard to evolve because developers know what should change, but not where the behavior lives.
- It introduces Harness Handbook, a behavior-centric repository representation synthesized from static analysis plus LLM-assisted behavioral structuring, linking runtime behaviors to source locations.
- It also adds Behavior-Guided Progressive Disclosure (BGPD), which guides coding agents from high-level behavior descriptions to likely implementation details and verifies candidate locations against current source.
- On modification requests from two open-source agent harnesses, handbook-assisted planning improved behavior localization and edit quality while using fewer planner tokens.
- Reported gains include +10.0 and +18.9 points in plan win rate, while reducing token use by 12.7% and 8.6% on Codex and Terminus-2 requests.
Related event: Harness Handbook Maps Agent Behavior to Source Code(2 posts)→
More from coding & agent
- Looking for a classifier of software engineering task shapes to pick models per task — StewartalsopIII · 2026-09-11
- Steal this idea: prompt-to-hardware where agents assemble custom devices — paraschopra · 2026-09-11
- Model Is the Least Interesting Part: A Guide to Six Core AI Architectures from RAG to Multi-Agent — goyalshaliniuk · 2026-09-11
- Non-coder builds layered memory architecture: 20k tokens tracks a year of agent conversations — matteoianni · 2026-09-11
- Warp's six non-engineering teams all run on Linear and Claude Code — mon__lim · 2026-09-11
- 9-year backend dev: AI code isn't the problem, the rate of making a mess is — Sweaty-Landscape-561 · 2026-09-11