huangruiteng/loopx

The open, provider-neutral, stateful control plane for long-horizon agents.

Runs on top of any agent harness — Codex App, Claude Code, Cursor, dsh, or your own — providing long-horizon state, semantic decisions, governance, recovery, and human-agent collaboration. Objectives, gates, todos, evidence, quota, and handoffs stay stable while the harness executes bounded turns.

Public website · Docs · Developer Book · Try LoopX · See real loops · How it works · User manual · 简体中文

把会干活的 Agent，接成可管理、可复盘、可持续改进的数字员工。

Open and provider-neutral, LoopX is a lightweight state kernel and local-first control plane for loop engineering. It runs on top of different agent harnesses rather than replacing them, providing the long-horizon state, semantic decisions about what happens next, governance, recovery, and human-agent collaboration that keep long-running work reviewable, restartable, and easier to hand off across turns, tools, and agents.

Loop engineering for long-horizon AI agents and peer agent teams.

Learn LoopX

• Developer Book - the curated bilingual path from control-plane foundations to project onboarding and developer contributions. 中文版 · English
• Getting started - install, connect a project, and run your first governed loop. Guide
• Docs - the full reference and operations site. LoopX Docs

Why LoopX

An agent can finish a task in one session. Long-running work is harder: objectives change, owner decisions appear, evidence goes stale, agents hand work to peers, and a scheduler can keep spending after no useful transition remains. Chat memory and a timer are not enough to govern that.

LoopX keeps the durable control state in one compact layer:

objective / issue / project
   │
   ▼
LoopX state: objective + gates + todos + scope + evidence + quota
   │
   ├─ human judgment needed? ── yes ─▶ ask a concrete question and wait
   │
   ├─ safe fallback available? ──────▶ run one bounded agent slice
   │
   ▼
Codex / Claude Code / Cursor / shell agent executes one turn
   │
   ▼
write evidence + handoff + next todo ─▶ quota decides the next tick

Agent runtimes execute the work. LoopX governs the state that lets engineering, research, discovery, and operations loops continue across runs. It is not another agent framework or a provider-specific orchestration runtime.

A useful mental model is an agent-native Kanban for long-running work. Cards carry identity, authority, evidence, and continuation. Moves are validated operators such as claim, gate, monitor, and writeback. The board is a projection; LoopX state remains the source of truth.

Registered agents are peers. Claims, leases, task boundaries, capabilities, and typed continuation decide who acts next; no durable leader identity is required.

LoopX is useful when you run:

• multi-day engineering, research, benchmark, or experiment objectives;
• issue and PR loops that must preserve scope, evidence, and review state;
• recurring heartbeat or monitor work;
• projects with owner, safety, publication, or private-data gates;
• peer-agent teams where ownership, leases, and handoff matter;
• creator, research, or operations workflows whose progress must remain legible to a non-engineering operator.

LoopX is not an autonomous production controller. Dangerous permissions, publishing, production writes, and final ownership stay with the human.

Evidence

These are not one-turn demos. The public OpenViking contribution sequence and the redacted, owner-run Auto ML showcase each span 200+ hours of elapsed loop lifetime across many bounded turns, decisions, and evidence updates. Elapsed lifetime is wall-clock project time. It is not 200 hours of continuous model execution or a claim of unattended production autonomy. Open each visual to inspect the public-safe graph, evidence branches, and decisions preserved across turns.

Open-Source Issue Fix

200+ hour public contribution arc: PR delivery and reusable fix knowledge evolve together.

LoopX's creator uses this path as an OpenViking contributor. The represented public contribution sequence spans more than 200 elapsed hours from its first PR creation to the latest represented review or update. The Issue-Fix capability keeps rolling repository context, revision-stamped fix knowledge, and reviewer-facing preferences separate; linked PRs plus current checkout source and tests remain authoritative.

Auto ML Experiment

Redacted owner-run showcase: a 200+ hour experiment arc keeps hypotheses, matched evidence, invalid lineages, running replicates, and promote/stop gates visible in one graph.

The redacted public-safe graph preserves decision lineage across that 200+ hour elapsed window. It is an owner-run showcase, not a claim of continuous compute, independent reproduction, a production result, or company or employer endorsement. The redacted image is not sufficient to reproduce the underlying experiment independently.

Auto Research

Reproducible public KNN demo: proposer, executor, and evaluator/promoter agents iterate in parallel while todo, quota, evidence, and targeted wake remain visible.

This screenshot comes from LoopX's built-in exact-KNN demo. The public task, editable and protected files, deterministic CPU evaluator, and dev/held-out commands all live in this repository. Follow the showcase walkthrough or the command path to reproduce the workflow; it is a demo result, not a production research claim.

Used In Real Projects

• Independent user · >13h C++ accuracy run. The user reported that a multi-stage task stayed aligned, triggered public research, adopted a public code-memory tool, and improved final precision. Read the evidence boundary.
• Independent user · 4d unattended run. The user reported four days without human intervention, useful ongoing work, and a periodic report surface. Read the redacted case.
• Independent user · 7 merged PRs. A LoopX-attributed Engine refactor is visible in a public issue and seven merged PRs; attribution and the reported 1B+ token scale remain user reports. Inspect the case.

These are the three strongest current cases, not the full inventory. Browse the complete Showcase catalog for contributor cases, creator dogfooding, reproducible demos, and explicit evidence-strength labels.

More inspectable surfaces:

• the public homepage for the product narrative, quick start, and long-running evidence;
• the complete Showcase catalog and its bilingual hosted index;
• the cross-runtime implementation review demo;
• the public user manual.

Try LoopX

Requirements: Python 3.11+. Use an active Python environment whose console scripts are on PATH; macOS and Linux use a POSIX shell, while native Windows uses PowerShell 7. Git is only needed for contributor clone/canary workflows. The package has no runtime dependencies outside the standard library.

Install from PyPI without cloning:

python3 -m pip install --upgrade loopx
loopx workflow-skills --install
loopx doctor

On native Windows PowerShell 7, use the same PyPI release without a POSIX compatibility layer:

py -3.11 -m pip install --upgrade loopx
loopx workflow-skills --install
loopx doctor

Restart your agent host after first install so it reloads the workflow skills. See Installing LoopX for pipx, host command surfaces, native Windows checkout installation, upgrade, rollback, uninstall, and the archive fallback.

Then connect from your project root:

cd /path/to/your-project
loopx connect
loopx status

If the project has not been initialized and connect tells you state is missing, use the guided path:

loopx start-goal --guided --project . --goal-text "Your long-running objective"

LoopX should reuse existing state rather than overwrite it. Keep .loopx/, .codex/goals/, and .local/ ignored.

Start From Your Agent

The exact, copy-ready setup messages and host recovery paths live in Getting Started. Host integrations can inspect the Codex App host command registry contract, the Codex CLI packaged install path, the Claude Code adapter, or the KunlunCode native Goal adapter.

For custom runners, start with the minimal custom runtime example (python3 examples/custom-runtime-minimal-cli-turn-smoke.py), then the full Embed LoopX in Your Agent Runner guide and the worker bridge install contract. The core tick is deliberately small:

loopx quota should-run      # should this registered agent act now?
loopx todo claim            # who owns this slice?
loopx todo update           # what changed?
loopx refresh-state         # what should the next turn see?
loopx quota spend-slot      # account for a completed, validated slice

First-Run Feedback

If LoopX works for you, a one-minute public issue helps us learn what a real first run looks like. It is optional, contains no telemetry, and should not include logs, paths, credentials, internal project names, or goal contents:

• First-run feedback
• Usage story for longer runs

loopx first-run-report prints the same prefilled link locally without sending anything.

A successful connection has:

• loopx doctor passing;
• .loopx/registry.json and a projected active goal state;
• loopx status showing the current objective, concrete user gate, and next agent todo;
• a visible loop driver or an exact activation instruction;
• local runtime state ignored rather than committed.

Clone-based install is only for contributors who want the live canary wrapper:

git clone https://github.com/huangruiteng/loopx ~/loopx
~/loopx/scripts/install-local.sh
loopx doctor

Capabilities

LoopX keeps the architecture explicit so the same governed outcome can survive a change of agent harness or external provider. The terms describe different boundaries rather than interchangeable kinds of plugin:

Host declarations such as --available-capability shell describe observed execution support. They are runtime capacities in this product map, not product capabilities and not permission grants. The effective capability still applies its own policy and authority checks before proposing a transition.

Core Control-Plane Promises

The Kernel folds its mechanics into five questions. Each question delivers one product promise on top of any agent harness: objective → long-horizon state; next → semantic decisions; human judgment → human-agent collaboration; evidence → recovery; continuation → governance.

Control-Plane Surface

The shipped primitives include lifetime goals, concrete user gates, audited safe fallbacks, peer todo ownership, quota and steering, compact run history, evidence-backed handoff, a read-first management surface, project-level value signals, and public/private boundary checks.

Product Capability Paths

Capabilities turn those generic primitives into outcome-owned work lanes. Start from the outcome, then inspect the current registered implementation and its write boundary:

Run loopx capability list --format json for the authoritative catalog in the installed release. A capability detail reports its user value, maturity, provider readiness, entry commands, write boundaries, protocols, and durable validation. Browse the human-readable capability index to choose by outcome; use Extensions and Capabilities when installing or building a provider.

Runtime Responsibilities

The execution path is Agent -> Capability -> Provider; the control path returns Provider readback -> Capability transition -> Kernel. An extension is how an optional provider is packaged and managed, not another control-plane owner. See Architecture and Extensions and Capabilities.

Advanced Paths

The first useful loop does not require every optional surface. Add these only when the work needs them.

Inspect the current goal's read-only capability catalog before enabling an advanced path:

loopx configure-goal --goal-id <goal-id>

Without --execute, this reports current/default state, fit, boundaries, and copyable commands without changing project state.

Presets and Auto Research

Safe presets cover daily triage, changelog drafts, and PR watching. The one-command research path coordinates proposer, executor, and evaluator/promoter roles while keeping quota and evidence visible. See the beginner preset guide and Auto Research command path.

loopx preset list
loopx preset show daily-triage

Preset inspection is read-only. For a connected recurring goal, loopx ready-score --goal-id <goal-id> --agent-id <agent-id> reports whether the loop is ready to run repeatedly.

Governed Turns

LoopX can generate one pure, bounded turn decision from a validated receipt, fresh quota state, and a provider-neutral budget. The current Codex CLI quickstart and activation contract are documented in LoopX Turn for Codex CLI.

Explore Graph and Harness

Explore is supported, optional, and default-off. It works best when a task has a measurable offline evaluation, baseline, treatment, and guardrails; it is not a substitute for production approval. Start with the Explore capability and its Lark presentation mapping.

Review Agent Work

Use loopx review-packet for a compact owner-facing view of decisions, evidence, validation, and unresolved gates. The intelligent management surface describes the operator model; the project-level reward model describes conservative value signals across output quantity, quality, token cost, and user attention cost.

For one concrete peer workflow, see the cross-runtime implementation review demo: Claude implements and Codex reviews while LoopX keeps ownership, evidence, quota, and handoff explicit.

App and Projection Paths

• Local read-first UI: dashboard guide
• Public product overview: public homepage
• Documentation portal: hosted docs
• Feishu/Lark projection: Lark Kanban adapter
• Generic host integration: integration guide
• Custom multi-agent runner: minimal custom runtime example, then custom runner integration

Optional projections make state easier to inspect; they do not become the source of truth.

Operating and Recovery

Start daily inspection with:

loopx status
loopx history --goal-id your-project-goal
loopx quota should-run --goal-id your-project-goal

Automatic turns must check quota first and append spend only after validated writeback. Quiet skips, preflight failures, and dry-run previews do not spend. When a user gate blocks one lane, a separately audited safe fallback may continue, but it must not bypass the gate.

Peer agents use loopx todo claim before delivery and loopx todo update after validation so ownership and evidence remain visible.

Scheduler cadence follows quota should-run.scheduler_hint; installed Codex App automations acknowledge the current hint through the returned ack_hint.cli_args. Collision recovery, monitor semantics, self-repair, and the exact operator commands are maintained in Getting Started, Quota Allocation, and Long-Task Cadence Policy.

Before publishing public docs or examples:

loopx check \
  --scan-path README.md \
  --scan-path docs/ \
  --scan-path examples/

Current Technical Directions

LoopX has three active strategic programs plus an architecture and research incubator. These are direction signals, not delivery promises; main, released artifacts, and stable reference contracts remain the source of shipped truth.

• Long-Horizon Benchmarks and Evidence: reproducible capability evidence and controlled mechanism research across complementary benchmark environments. Direction tracker
• Operator Surface and IM Integration: an operator workspace, session records, and bounded collaboration surfaces, currently incubating on a dedicated integration branch with @maxliux5 as implementation lead. Direction tracker
• Shared Goal Authority and Cross-host Coordination: provider-neutral coordination for explicitly shared goals, with NoKV as an unpromoted provider candidate rather than a new control-plane authority. Direction tracker
• Architecture and Research Incubator: Effect Program hardening, TypeScript parity migration, hierarchical stride, research exploration, human attention, artifact lifecycle, and memory utility work at explicitly different maturity levels. Direction tracker

Read the canonical Technical Directions map for stages, promotion gates, contributor-safe cuts, and ownership boundaries. Use the pinned GitHub Discussion for community discussion. Core control-plane reliability continues as the shared foundation beneath these programs.

Advanced Documentation

Start with the path that matches your current task. Use the hosted documentation portal for the published docs site; the documentation index remains the complete source map. This list stays selective; each category index owns its deeper documents and versioned protocols.

Use and Operate

• Getting Started: install, connect, diagnose, daily workflow, heartbeats, dashboard, development, and commands.
• User Manual: public onboarding, concepts, FAQ, and selected cases.
• Operations: goal continuation, todo, cadence, attention, and authority workflows.
• Quota Allocation and Heartbeat Automation Prompt: scheduler eligibility, spend, and scheduled continuation.
• Dashboard and Status Data Contract: operator-facing state and projection contracts.
• Release Readiness: install/update paths, compatibility gates, release notes, and safe-to-depend-on surfaces.

Understand the Control Plane

• Architecture: lifetime-goal invariant and kernel.
• State Interaction Model: actors, stores, interaction contract, and writeback.
• Concepts: reusable routing, gate, evidence, projection, and planning patterns.
• Product Foundations: Loop Engineering principles, project-level reward, and reward-style replanning.
• Product Vision: the broader Loop Agent direction.

Integrate and Extend

• Integration Guide
• Minimal Custom Runtime Example
• Custom Agent Runner Integration
• Integrations: runtime, host, collaboration, and external-system adapters, including worker bridge and Lark.
• Extensions and Capabilities

Build and Review LoopX

• Developer Guide: contributor workflows, benchmark development, documentation layout, and quality gates.
• Reference and Protocols: stable contracts and versioned implementation protocols, including host command and reward memory architecture.
• Control-Plane Developer Course: nine Chinese, code-led lectures.
• Testing and Quality: validation layers and risk-based checks.
• Public/Private Boundary: safe fixtures, examples, evidence, and publication.

Inspect Outcomes

• Showcase Catalog: public-safe cases and evidence labels.
• Research and Evidence: benchmark investigations and source-backed findings.
• Update Notes: public-safe progress notes.

Project and Community

• Current Technical Directions
• Project Governance
• Contributing and Contributor Tasks
• Authors and Contributors
• Project History
• Name and Marks
• Ecosystem Adoption - integrations, sampling, and derivatives we observe and track

Partner Projects

LoopX welcomes collaboration with other open-source projects to build the long-running agent ecosystem. Our confirmed partners include:

• OpenViking - Self-evolving context database for AI agents
• NoKV - AI native distributed file system

Community and Feedback

LoopX is already running real long-running agent goals and is under active development. The most useful feedback comes from real long-running agent projects: where the control plane helped, where it felt heavy, and which gates or handoffs disappeared from view.

• Use GitHub Issues for reproducible bugs, install problems, and feature requests.
• Open PRs for docs fixes, showcase writeups, and small public-safe examples.
• Join the Discord community, or use Lark or WeChat below.

See Support for channel routing, service boundaries, and official publication sources.

Contributing

External contributors should start with Contributor Tasks for public, claimable work and Contributing for setup, validation, and boundary rules. Project roles and public history are recorded in Governance, Authors and Contributors, and Project History.

LoopX keeps local active state separate from the public repository. Do not commit .loopx/, .codex/goals/, live ACTIVE_GOAL_STATE.md, raw benchmark traces, credentials, private logs, or operator artifacts.

Current Status

The v0.4.x line is a usable local control plane for long-running agent work and is entering broader adoption. It is not a full agent platform, an agent runtime, or an autonomous production controller.

Today LoopX ships a durable state kernel for goals, typed todos and decision scopes, peer claims and leases, evidence and writeback, quota-aware scheduling, and cross-turn continuation. Guided start, recurring heartbeat, isolated Codex CLI turns, evidence-backed Issue-Fix admission, optional Explore and auto research paths, public validation canaries, and a read-first multi-project dashboard build on that shared control state.

Support levels remain explicit. The state and CLI contracts are the stable center; several host integrations and advanced paths are optional, default-off, or experimental. LoopX does not grant credentials, approve destructive or production actions, publish on a user's behalf without authorization, or turn an unverified run into evidence of success.

Current investment is organized through the Technical Directions map: long-horizon benchmark evidence, operator surface and IM integration, shared-goal cross-host coordination, and an explicitly staged architecture and research incubator.

Star History

License

Apache License 2.0 beginning with v0.4.8. See LICENSE and NOTICE. Releases through v0.4.7 remain under their original MIT terms; the historical text and notice are preserved in LICENSE-MIT. The licensing policy explains the version, contribution, patent-grant, and open-core boundaries.