Alibaba Cloud Software / SaaS China Translated from Chinese

ECLICKTECH (易点天下)

IT Operations Automation

ECLICKTECH's operations team had to manage a multi-cloud, multi-product environment that supported several lines of business, but its observability data, alerting systems, and AI tooling were fragmented across separate platforms and personal workspaces. That made end-to-end troubleshooting slow, prevented the team from turning individual agents and skills into shared organizational assets, and raised governance concerns around multi-agent execution, least-privilege access, auditability, and human fallback as the company tried to scale AI-assisted operations.

Read the original story Explore all use cases

connects more than 600,000 media outlets globally

covers 200+ countries and regions

multi-account ECS metric onboarding and aggregated delivery cycle was shortened from days to hours

serves more than 10,000 enterprises and brands expanding overseas

Reported impact

Business outcome

ECLICKTECH replaced fragmented monitoring and individually managed AI tooling with an organization-wide intelligent operations loop. The unified observability foundation improved troubleshooting efficiency by giving teams one view from infrastructure to application, STAROps reduced repetitive work by turning multi-account ECS metric onboarding from a days-long manual process into an hours-long natural-language workflow, and AgentCore plus AgentLoop gave the company reusable agent governance, security scanning, auditability, and continuous evaluation for scaling multi-agent operations without duplicating skills.

Solution

ECLICKTECH rebuilt its intelligent operations architecture on Alibaba Cloud as a three-layer stack spanning business systems, a shared AI and data middle platform, and intelligent multi-cloud management infrastructure. Cloud Monitor 2.0, Managed Service for Prometheus, and Log Service SLS aggregate metrics, traces, logs, events, and alerts from multiple cloud products into one end-to-end observability view. AgentCore then serves as the control plane for operations agents and skills: it registers hosted and managed agents, routes requests with a Leader Agent, orchestrates multi-agent tasks, and connects through MCP to ContextDB so historical cases and operations knowledge can be reused. STAROps exposes those capabilities through natural-language workflows for tasks such as multi-account ECS metric onboarding, while AgentLoop records call chains, sessions, token usage, execution traces, audits, and evaluation results so the team can govern and continuously improve agent behavior.

Data flow

Metrics, traces, logs, events, and alerts from ECLICKTECH's multi-cloud business systems and ECS accounts are aggregated through Cloud Monitor 2.0, Managed Service for Prometheus, and Log Service SLS into one observability plane. AgentCore registers agents and skills, lets a Leader Agent route requests, and connects to ContextDB over MCP for historical knowledge. STAROps exposes those capabilities through natural-language workflows, while AgentLoop records calls, tokens, and audit results for governance and improvement.

Solution architecture

6 components · 4 layers

Ingest
- Cloud Monitor 2.0 Aggregates multi-cloud operations telemetry into the shared observability base used for end-to-end monitoring.
Storage
- Managed Service for Prometheus Collects and aggregates ECS and other infrastructure metrics for cross-account monitoring and delivery.
- Log Service SLS Stores logs and event data so operators and agents can investigate incidents from a unified view.
Orchestration
- STAROps Provides the natural-language operations interface for daily automation tasks and efficiency gains.
- AgentCore Manages agent and skill registration, routing, orchestration, and lifecycle governance across the operations estate.
Governance
- AgentLoop Captures execution traces, audits risk, and supports continuous evaluation of agents and skills.

Architecture clues

A Leader Agent handles scenario recognition, intelligent routing, and multi-agent task orchestration inside AgentCore.
AgentLoop records call chains, sessions, token consumption, and execution traces for audit and evaluation.
ContextDB is connected through MCP so agents can query historical cases and operations knowledge in real time.
Feishu bot connectivity and internal-system integration are part of the governed agent-management layer.
The overall AI architecture is organized into business systems, a unified AI and data middle platform, and intelligent multi-cloud management infrastructure.

Evidence from the source

Alibaba Cloud aggregates multi-cloud operations data through a unified observability foundation, uses AgentCore to govern and orchestrate agents and skills, and relies on AgentLoop for end-to-end observation, risk auditing, and continuous evaluation. (original: 阿里云以统一可观测底座汇聚多云运维数据，通过 AgentCore 统一治理和编排 Agent 与 Skill，并借助 AgentLoop 实现全链路观测、风险审计和持续评估，形成“观测—智能—治理”一体化的智能运维闭环。)

Based on AgentCore, the company created a unified intelligent-operations entry point, connected multiple agents with internal systems and a Feishu bot, and applied skill security scanning so personal AI productivity practices could become organization-level agile assets. (original: 基于 AgentCore 构建了智能运维统一入口，实现所有运维 Agent 和 Skill 的全生命周期协同管理。完成了多 Agent 与内部系统的打通、飞书机器人对接、Skill 安全扫描等关键集成，将个人的 AI 提效实践真正转化为组织级的敏捷资产，有效避免了 Agent 和 Skill 的重复建设。)

By combining Cloud Monitor 2.0, Prometheus, and Log Service SLS, the company aggregates metrics, traces, logs, and events into a unified view that provides end-to-end observability from infrastructure to the application layer. (original: 结合阿里云云监控 2.0、Prometheus 和日志服务 SLS 等云产品，将指标、链路、日志、事件汇聚于统一视图，提供从基础设施到应用层的全链路、端到端的统一观测能力，打破多云环境下的数据孤岛。)

The company says its headquarters are in Xi'an and it has subsidiaries in Beijing, Shanghai, Guangzhou, Hangzhou, Chengdu, Shenzhen, Hong Kong, South Korea, Japan, Germany, the United States, India, BVI, and the Cayman Islands. (original: 公司总部位于西安，并在北京、上海、广州、杭州、成都、深圳、香港、韩国、日本、德国、美国、印度、BVI、开曼等国家和地区设有子公司。)

The company's overall AI architecture is divided into three layers: the business systems layer, the unified AI and data middle platform layer, and the intelligent multi-cloud management infrastructure layer, forming a closed loop in which the business calls downward and data feeds back upward. (original: 公司整体 AI 架构自上而下分为三层：业务系统层、统一 AI 中台与数据中台层、智能化多云管理基础设施层，形成业务向下调用、数据向上反馈的闭环体系。)

With the STAROps intelligent operations platform, natural-language capabilities are used for daily operations tasks, and multi-account ECS metric onboarding and aggregated delivery were shortened from days to hours. (original: 基于 STAROps 智能运维平台，借助自然语言能力完成日常运维任务。例如，多账号下的 ECS 指标批量接入 Prometheus 和聚合投递，以往需要人工逐台配置，工作周期从天级缩短到小时级，准备率和操作效率均大幅提升。)

Transparency Evidence & methodology 75/100 reference value · 93% confidence · 2 sources

Review how this reference was scored and trace each normalized claim back to its documented source.

Scoring methodology

How scores are calculated

Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.

Customer reference value75/100

Problem clarity20/20
Measures whether the customer's operational problem is specific enough to compare.
Implementation depth25/25
Rewards named components, data flow, integrations, and architecture detail.
Outcome specificity20/20
Measures whether the reported benefit is concrete rather than a generic claim.
Quantified impact0/25
0 metrics describe measurable impact.
Deployment scale10/10
Rewards evidence of a live or scaled deployment and meaningful integration scope.

Evidence confidence93%

Source authority20/25
The main evidence is an official vendor-published customer story.
Claim traceability25/25
12 of 12 key fields link to a named source.
Direct evidence20/20
6 direct source quotes retained.
Date provenance10/10
The publication date comes from the source page or publisher.
Source transparency10/10
2 source references document publisher, URL, authority, retrieval, and supported fields.
Implementation specificity8/10
Rewards source-backed components, data flow, quantified impact, and a specific outcome.

Traceability

Sources & provenance

Field coverage shows which facts each source supports. Dates are shown only when documented by the source.

Alibaba Cloud
易点天下基于阿里云智能运维产品构建统一可观测与智能体治理体系

First-party
https://www.aliyun.com/customer-stories/internet-and-info-tech-2026-eclicktechsre

Source type
Vendor-published customer story
Supports
- Business problem
- Solution
- Products
- Solution components
- Data flow
- Integrations
- Business outcome
- Metrics
- Architecture
- evidence quotes
ECLICKTECH
践行社会责任走进易点天下

First-party
https://www.eclicktech.com.cn/html/company/931.html

Source type
Company profile
Supports
- Customer
- Region
- Company size
- Metrics

Date evidence

Published on source page Oct 8, 2026 · Alibaba Cloud · First-party
Open date source

Not separately sourced Industry. Treat these normalized attributes as lower-confidence until a dedicated source is added.

Similar use cases

See all like this

Alibaba Cloud Oct 12, 2026 ECLICKTECH (易点天下) Software / SaaS China ECLICKTECH needed to industrialize AI-driven advertising operations, but media-delivery workflows were still long, manual, and heavily dependent on senior operators for anomaly diagnosis and creative adaptation. As more… AI covers about 50% of core execution work IT Operations AutomationMarketing Personalization AdClaw · AgentCore · AgentLoop

Alibaba Cloud Jul 28, 2026 Zhuiyin Technology (追因科技) Software / SaaS China Zhuiyin Technology had already built an agent for cross-border ecommerce analysis, but once the agent was exposed to customers, the team discovered three hard problems: results were hard to judge objectively, reasoning… inference quality improved 20% Application RefactoringIT Operations Automation AgentLoop · Trekin Agent

Alibaba Cloud Jul 22, 2026 Chanjet Software / SaaS China Chanjet needed user-centric observability and faster incident handling for a multitenant, multi-center SaaS platform serving millions of small businesses. Its traditional monitoring stack could not detect… 10-minute emergency containment target IT Operations Automation Application Real-Time Monitoring Service (ARMS) · Cloud Monitor CMS · Log Service SLS