Scattered agent capabilities
Tools, scripts, MCP connectors, and browser automation live in separate personal setups.
Your first digital employee starts here
Choose, use, teach, reuse, and manage digital employees across desktop workspaces and enterprise platforms.
Digital employee marketplace
pnpm test --filter webQianGong connects agents, Skills, workspaces, and operation records into one lifecycle so expertise can be used, reused, and governed.
Tools, scripts, MCP connectors, and browser automation live in separate personal setups.
Experts know the workflow, but their knowledge rarely becomes a reusable capability.
Teams need to know who used which employee, what happened, and what result was produced.
Some jobs require a desktop environment, while others belong in cloud workspaces.
Digital employees, Skills, clients, permissions, and records need one management path.
From defining employees to teaching, Skill reuse, marketplace distribution, and operation records, QianGong turns one-off work into managed assets.
Set the role, task boundary, entry point, and use case.
Filter by development, office work, knowledge search, data analysis, and business operations, then add the right employee to your workspace.
Run local tasks in the desktop client or lightweight tasks in the cloud workspace, with process and result records.
Use CLI, MCP, browser demonstrations, and spoken context to turn real workflow knowledge into Skill drafts.
Users, teachers, individual operators
Local execution, desktop environment access, marketplace, and personal employees
Business experts, implementers, developers
CLI/MCP/browser demonstrations, spoken context, and Skill drafts
Business users and implementers
Record page operations, voice notes, request summaries, and teaching uploads
Enterprise employees
Choose digital employees, run lightweight tasks, and create cloud records
Admins, IT, AI leaders
Manage digital employees, Skills, clients, and operation records
Teachers, developers, admins
Reusable capability assets, employee binding, and organization sharing
Review changes, summarize risks, and suggest fixes
Read logs and identify likely error paths
Business experts do not need to start with an agent framework. QianGong turns a demonstration, command, or tool connection into a Skill draft that can be reviewed, tested, and reused.
Turn commands, parameters, outputs, and risk rules into CLI Skills.
The first launch focuses on visibility for employees, Skills, clients, and operation records. Deeper permissions, approvals, and runtime governance are clearly marked as planned.
The execution core follows the Pi Agent direction, while the product layer presents it as digital employees and Skills.
Question-answering first, weak execution and management
A lifecycle around digital employee tasks, Skills, and operation recordsStrong workflow automation, weaker teaching and intelligence
Demonstration teaching and Skill reuse first, richer replay laterDeveloper-facing and concept-heavy
Productized as digital employees, Skills, and marketplacesPersonal use first, weak enterprise governance
Covers both desktop usage and enterprise managementReview changes, find risks, suggest fixes
Code review Skill / Git SkillEngineering teamsDraft requirements, architecture, and delivery docs
Documentation SkillPMs / ArchitectsRead logs, attribute errors, produce investigation paths
Log attribution SkillEngineering / OpsSearch policies, documents, and knowledge bases
Retrieval SkillEnterprise employeesLearn query and form workflows from browser systems
Browser teaching SkillBusiness expertsSummarize meetings, actions, and risks
Meeting summary SkillManagers / Delivery teamsA normal assistant is usually chat-first. QianGong is built around digital employees, Skills, workspaces, and records that can be reused and governed.
A digital employee is the role users choose for a task. A Skill is a reusable capability that can be bound to one or more employees.
Many tasks need local files, CLI tools, MCP connectors, or the user's machine environment. The desktop client is the local work and teaching entry.
Private deployment is a product direction for enterprise digital employee teams. This launch page does not promise unfrozen delivery details.
Browser recording is a teaching input for authorized demonstration steps, spoken context, and necessary workflow context. Sensitive data boundaries require formal implementation design.
Business experts can teach through browser demonstrations and spoken context. Developers can add more structured capability through CLI and MCP.
The launch page presents the loop of CLI, MCP, browser demonstration, and human confirmation. Production details follow official releases.
The first launch emphasizes who used which entry, which employee, what was done, and what result was produced. Audit depth expands with governance features.
Turn expert know-how into reusable Skills and help your organization build its own digital employee team.