Possible complementary direction: LongHorizon-Harness for sustained agent tasks

Hi, I'd like to suggest a complementary direction for `BitFun`.

Project:
https://github.com/AMAP-ML/LongHorizon-Harness

Your project focuses on an agent harness and orchestration layer.

LongHorizon-Harness focuses on carrying one task across many rounds without state drift. It keeps the goal and verified progress in durable state, starts each execution round with a fresh context, and uses an independent Auditor before accepting progress.

Why this may be useful here:

- A short workflow note could show how to avoid context drift and premature completion in a task that spans many rounds.
- it adds an outer task-lifecycle layer without claiming to replace the project's existing agent or runtime

If this direction fits, I can prepare a small documentation or example contribution following the repository's conventions.

I am part of the team behind LongHorizon-Harness, so I wanted to disclose that connection.

Paper:
https://arxiv.org/abs/2608.01964

Project page:
https://lh-harness.pages.dev

The issue names no repository file, test, or entry point. Start by reviewing the repository's documentation and example conventions, then decide whether the proposed LongHorizon-Harness workflow fits. Done would be a scoped, accepted documentation or example contribution, but the issue does not define its contents.

Rédigé par le modèle d'indexation à partir du texte de l'issue.