Notebook · 9 items
Harness² · repo guide
Sunil TiwariPublished Oct 27, 2026 · updated Oct 27
Project · note + canvas01 · Architecture overview8 min read · Oct 26- 08 · Setup and running · This chapter takes you from an empty Linux machine to a judged, reported Harness² benchmark run. Harness² ships seven console commands (harness2-demo, harness2-setup, harness2-check, harness2, h…Project · note + canvas08 · Setup and running35 min read · Oct 26
- 03 · One task, end to end · This chapter follows one benchmark task through the whole Harness² run lifecycle, at commit 1cfe254. A sweep turns a benchmark split into a seeded, ordered stream of (domain, task) pairs. It…Project · note + canvas03 · One task, end to end40 min read · Oct 26
- Harness² keeps three things separate: the method (recursion, prompts, selection, library), the substrate (the agent framework a solver rollout runs under: OpenCode or Codex CLI), and the benchmark (a BenchSpec value plu…Note07 · Executors and benchmarks47 min read · Oct 27
- In short. Harness² compares two ways of improving a task-executor's harness (a parallel mode and a sequential mode) and the comparison is only fair if both spend exactly the same. A small ledger records every executor r…Note06 · Budget, selection and judging37 min read · Oct 27
- Harness² improves a harness by handing an "improver" model (also called the proposer, the editor or the selector, depending on its job) a folder of evidence and letting it read and edit files there. This chapter covers…Note05 · Evidence and safety47 min read · Oct 27
- Summary. In Harness², a "harness" is a small tree of text files around a fixed agent framework (OpenCode or codex-cli). An improvement step may change only these files, and the eight component slots c1..c8 decide which…Note04 · The harness tree42 min read · Oct 27
- Harness² is one Python package, harness2, split into four subpackages: core (the method), substrate (the agent frameworks that run a solver), benchmarks (the three benchmark adapters) and cli (the command-line entry poi…Note02 · Code structure48 min read · Oct 27
- A chapter-by-chapter guide to google-research/harness2 (Harness²: recursive agent harnessing). Pinned to commit 1cfe254, so every code link points at exactly the code described. · Reading order · | # | Chapter…Note00 · Start here9 min read · Oct 27
Make a page like this. Free.Notes and diagrams side by side, written by you or by Claude Code and Codex. Publish it in one click.
Make your own, free