AstarterPreparing the research scene0

Astarter AI · LensWave AI Trading OS · Trading OS

Astarter
中文
Astarter
Technology Trading OS Risk framework EcosystemDocsBlog
Open the AI workspace

LENSWAVE · ASTARTER AI TRADING OS

Intelligence in concert
Strategies in motion

Astarter connects compute, data, agents and applications for open finance. LensWave leads with AI trading; AI Portfolio explores allocation; a future Onchain Bot will connect onchain trading intent. Research informs decisions, rules govern action, and experience builds expertise.

Explore the technology

LensWave Collective

Different perspectives
Shared evidence

Bring market activity, capital, events and risk into traceable research. Different roles examine evidence and counter-evidence around the same opportunity, making judgments and disagreements clearer. Saving, reuse and recovery have bounded local validation; live collaboration across models remains in development.

Strategy Lineage

Ideas become rules
Versions stay traceable

Record entry, exit, invalidation and parameter dependencies in a reviewable strategy version. Input/output rules and validation checks have been tested; strategy graph compilation and the complete runtime remain in development.

LensWave Trading Core

Rules guide execution
Facts explain outcomes

Each approval binds a specific version; each instance respects its account and shared capital constraints. LensWave Trading Core connects checks, reservations, orders, fills and fees. Unknown outcomes require reconciliation of the original operation. Components have local simulation evidence; continuous live execution remains unaccepted.

LensWave Learning Engine

Review the experience
Then build the dataset

A strategy revision, risk explanation or corrected review can be the start of specialist knowledge. The learning engine aims to select authorized cases and build trading-model capability through offline training and independent evaluation. This remains an R&D goal.

LensWave Trading Model

Expertise through training
Progress through evaluation

LensWave Trading Model targets defined tasks: clarifying intent, finding missing rules, checking citations, explaining risk and organizing reviews. Candidates must pass independent evaluation, shadow observation and controlled release. Completed specialist training and improved results have not yet been established.

Astarter AI Ecosystem

Built around trading
Connected to an AI ecosystem

Astarter organizes data, tools, agents and applications. LensWave focuses on AI trading; authorized and validated experience may later help improve specialist tools and services.

Evidence informs research
Discipline guides action

Astarter builds the AI ecosystem. LensWave brings its capabilities to trading. Each action leaves records that can be checked; with consent and validation, those records can support the development of specialist expertise.

Explore the Astarter ecosystem

BACKED BY CONVICTION

与长期建设者同行

ASTARTER × LENSWAVE

Ecosystem capability
Tested in trading

Settlement, compute, agents and applications form Astarter’s four layers. LensWave focuses on AI trading, AI Portfolio explores allocation, and Onchain Bot is a planned trading interface. They share research and tools while developing distinct permissions, risk controls and delivery requirements.

Ecosystem capability · Tested in trading

Four layers connect settlement, compute, agents and application services. AI DEX integration targets market data and execution; prediction markets inform event research; data markets supply licensed inputs; agent markets assemble expertise. These are integration goals, each subject to separate validation.
Infrastructure
Settlement

Accountable execution and service settlement

Data & compute
Compute

Match resources to licences and tasks

Agent coordination
Coordination

Coordinate roles, tools and permissions

Applications
Services

AI trading, portfolio allocation and onchain trading interfaces

LENSWAVE / AI TRADING OS

Research together
Execute with discipline

Connect research, configuration, approval and review from market evidence to strategy operation. AI proposes a reviewable plan; the trader confirms its version, capital and risk boundaries. LensWave Trading Core is designed to manage instances and orders within those limits. Current validation is primarily bounded local simulation.

LensWave Collective

Roles and interfaces defined

Different perspectives
Shared evidence

Bring market activity, capital, events and risk into traceable research. Different roles examine evidence and counter-evidence around the same opportunity, making judgments and disagreements clearer. Saving, reuse and recovery have bounded local validation; live collaboration across models remains in development.

Planned roles
18 logical roles
Responsibilities
Observe · argue · review · reflect
Research output
Sources · disagreements · versions
Current stage
Live collaboration in development

LensWave Trading Core

Limited components validated

Rules guide execution
Facts explain outcomes

Each approval binds a specific version; each instance respects its account and shared capital constraints. LensWave Trading Core connects checks, reservations, orders, fills and fees. Unknown outcomes require reconciliation of the original operation. Components have local simulation evidence; continuous live execution remains unaccepted.

Events and timing
Preserve what was knowable then
Order lifecycle
Separate partial fills from final states
Capital responsibility
Precise sizing · atomic reservations
Unknown outcomes
Retain reservations; reconcile first

Learning loop · Development goal Explicit consent → selection and correction → offline training → independent evaluation → shadow operation → controlled release. Validated capability returns to research; model upgrades never silently rewrite active strategies or account permissions.

Explore the technical loop

RISK FRAMEWORK / THREE LAYERS

Pursue the opportunity
Know when to pull back

Set limits per trade, reveal correlated exposure and enforce portfolio discipline. Three checks align strategies around shared boundaries, reducing or pausing new risk when limits are reached.

三层防线 · 同一套风险纪律从每次决策，贯穿整个组合

01

单笔约束

从可承受的风险出发，结合止损与成本约束每次配置。

02

关联敞口

识别同类资产的叠加暴露，避免多套策略押向同一风险。

03

组合防线

统筹组合风险，在回撤与亏损触线时收紧或暂停新增风险。

机会可以变化，边界始终在场

预设约束 → 联合核验 → 收紧或暂停

了解风控体系

Six research perspectives, one wider market view

  • MARKET
  • LIQUIDITY
  • MACRO
  • SENTIMENT
  • ON-CHAIN
  • RISK

Enter Astarter

Build the next chapter
of open finance

Use Docs to explore LensWave’s architecture, Trading Core and learning roadmap. Read the Blog for the flagship product’s place in Astarter’s broader AI ecosystem.

Choose your starting point

Find a strategyClarify goals, risk preferences and decision mode Enter the agent workspaceFollow the evidence across research perspectives Explore strategiesCompare mechanisms, conditions and exit rules

ASTARTER / INTELLIGENT DECISION WORKFLOW

Start with five questions

Five questions clarify your goals, acceptable limits and research pace, giving you strategy directions to compare.

  1. Define goals and research preferences
  2. Compare strategy and risk boundaries
  3. Set limits, validate and review
Find my strategy

Before you begin

Where should I start?

Astarter is the overall AI ecosystem; LensWave is its flagship AI Trading OS in development. Explore Docs and the Blog to understand the mechanisms and current progress. Launch APP will open when the product is available.

How do agents research together?

Bring market activity, capital, events and risk into traceable research. Different roles examine evidence and counter-evidence around the same opportunity, making judgments and disagreements clearer. Saving, reuse and recovery have bounded local validation; live collaboration across models remains in development.

How do I choose a suitable strategy?

Current evidence covers adapted grid and bounded DCA versions in local simulation. Catalog inclusion does not establish executability, and joint acceptance of two BOT flows is still pending. Simulation records cannot prove real fills or a live trading connection.

How do the three layers work together?

Trade limits constrain each allocation, correlation checks identify overlapping exposure, and portfolio rules reduce or pause new risk at loss and drawdown thresholds. Review actual conditions in strategy settings and records. Limits do not guarantee returns or a maximum loss.

How do I review my research and strategy?

Read evidence and disagreements in the workspace, then check triggers, parameters and risk responses in process records. Separate views, plans and confirmed results to identify the next improvement.