Tencent and universities release 29-page Harness Handbook for evolving agent harnesses
blaizedsouza · x · 2026-11-18
A 29-page PDF, "Harness Handbook: Making Evolving Agent Harnesses Readable, Navigable, and Editable," from Tencent HY LLM Frontier, Indiana University, University of Maryland, and others:
- Problem: agent capability = model + harness; production harnesses are large, tightly coupled, and behaviorally scattered, making behavior localization the central bottleneck of harness evolution.
- Method: Harness Handbook is a behavior-centric codebase representation auto-synthesized via static program analysis plus LLM-assisted behavioral structuring, linking each behavior to source code; paired with Behavior-Guided Progressive Disclosure (BGPD) that guides coding agents from high-level behavior descriptions to implementation details with verification.
- Results: on modification requests across two open-source harnesses, handbook-assisted planning improves behavior localization and edit-plan quality with fewer planner tokens; biggest gains on scattered implementations, rarely executed paths, and cross-module changes.
Project: ruhan-wang.github.io/Harness-Handbook
More from coding & agent
- Report: NSA and Wall Street execs tell Congress AI-found vulnerability patching caused outages — moyix · 2026-11-18
- Sysdig CISO: LLMs can rewrite the harness rules meant to govern them — massimosgrelli · 2026-11-18
- shadcn's agent-first UI SDK: ~50% fewer tool calls, ~75% fewer tokens, ~70% cheaper — shadcn · 2026-11-18
- Slite analyzed 50 knowledge bases: AI does 1 in 5 reads, 98.7% of docs unverified — femke_plantinga · 2026-11-18
- AI already reads 20% of knowledge base docs, but only 1.3% are verified — femke_plantinga · 2026-11-18
- oxlint hits 50 million weekly downloads, adding 10 million in three weeks — cnakazawa · 2026-11-18