12 Reusable Agentic Harness Patterns Distilled From Claude Code's Design
bibryam · x · 2026-10-10
The author distills 12 reusable harness patterns from Claude Code, arguing the model is only one part of a coding agent — the harness decides what the model remembers, which context it sees, how work is split, which tools it can use, and which actions need approval.
Memory and context
- Persistent instruction file: load a project-level file each session carrying build commands, architecture rules, and coding standards; costs maintenance as stale instructions mislead.
- Scoped context assembly: large repos must be filtered to task-relevant context.
Other categories
- Workflow & orchestration, tools & permissions, and automation patterns are covered in the full article.
Core thesis: a coding agent's ceiling is largely determined by harness engineering, not the model itself.
More from coding & agent
- Claude Code Pushes Urgent Security Update Across All Installations — umeshonai · 2026-10-23
- Claude cut a 2.6-hour video export to 20 minutes — but only after being told to rebuild the server — byhewar · 2026-10-23
- 3 flight-booking rules my AI agent checks that I'd never do by hand — Efistoffeles · 2026-10-23
- OpenJet claims Qwen3.8-27B coding runs 1.9x faster than llama.cpp on one RTX 3090 — JeremyCMorgan · 2026-10-23
- Building an autonomous SEO agent in Hermes with three skills: programmatic pages, Keywords Everywhere, DataForSEO — EXM7777 · 2026-10-23
- Building a sports betting AI agent with onchain execution and human approval by default: architecture and lessons — Dry-Comfortable-2678 · 2026-10-23