Sign InOpen Brain
GitHubGitHub RepoNeeds Review

huangruiteng/loopx

LoopX gives long-running coding-agent work durable goals, gates, ownership, evidence and quotas across bounded turns. It is useful when chat history and a scheduler no longer provide enough control.

GitHub · Trending today
Open Source Open MarkdownOpen JSON
Source Summary

LoopX **v0.4.x** is a local, agent-agnostic state kernel for Codex, Claude Code, Cursor and custom runners. It preserves **goals, gates, todos, evidence and quota** while agents execute bounded turns; two published trajectories span **200+ elapsed hours** each.

Practical Implication

Use it when work crosses sessions or agents and ownership, stop conditions or human approvals must remain explicit. Its quota check, todo claims, validated writeback and review packets offer a concrete harness pattern even if you do not adopt the project.

Agent-Ready Context
LoopX **v0.4.x** is a local, agent-agnostic state kernel for Codex, Claude Code, Cursor and custom runners. It preserves **goals, gates, todos, evidence and quota** while agents execute bounded turns; two published trajectories span **200+ elapsed hours** each.

Use it when work crosses sessions or agents and ownership, stop conditions or human approvals must remain explicit. Its quota check, todo claims, validated writeback and review packets offer a concrete harness pattern even if you do not adopt the project.

It is early software, and the long trajectories measure wall-clock project lifetime rather than continuous execution or independent reproduction. **Production writes and final authority stay with the human**, so LoopX is not an autonomous production controller.
Connected Context · Feed7 Judgment

LoopX turns prior guidance about durable memory, bounded work, evidence-based completion and human authority into a small local state kernel that can span different agent runners. It supports the case for explicit governance across long-lived work while narrowing the evidence: its 200-hour trajectories show persistence over project time, not continuous autonomous execution or independently validated reliability.

What Does Done Even Mean? Agents and Paperclip's Liveness Model - Dotta, PaperclipLoopX operationalizes the candidate’s evidence-and-authority model of completion through gates, validated writeback, review packets and explicit human final authority.Chained Recursive Language Models for Multi-Iteration ReasoningBoth preserve structured state across bounded fresh-context iterations; LoopX adds operational ownership, quota and approval controls around that persistence pattern.Why We Killed Our Multi-Agent Pipeline — Subbiah Sethuraman and Abhilash Asokan, ZS AssociatesTodo claims and bounded turns can support the candidate’s preferred model of clear end-to-end ownership with narrowly scoped delegated work, rather than a fixed specialist chain.affaan-m/ECCWhere ECC offers a broad cross-agent catalog, LoopX provides a narrower shared state and governance substrate that could coordinate selectively adopted skills without replacing their verification needs.
Context Map
agentcodingresearch#harness-engineering#multi-agent#agent-memory
Uncertainty
It is early software, and the long trajectories measure wall-clock project lifetime rather than continuous execution or independent reproduction. **Production writes and final authority stay with the human**, so LoopX is not an autonomous production controller.