Solutions

Three kinds of work,
one chain.

E-commerce operations, education-sector admissions, engineering teams — all of them decide often and lean heavily on experience. Each solution combines one or two of our three product lines: RPA, workflow and CRM. Use any one alone, or chain them together

Real data
Modelled from human expertise
Automated execution
Logged and traceable
RPAWorkflow · ChengguanCRM

E-commerce operations

Three things. Invoicing and reconciliation — RPA takes over invoicing on the national tax platform across multiple legal entities, and writes back to the ERP. Nobody clicks through them one by one any more. Restocking and clearance — Chengguan works out margin and ROI by store, style and SKU, then produces restocking recommendations, clearance lists and return-rate analysis. The rules come from your operators' experience, not a model's own judgement. Customers and channels — distributors, influencers and suppliers are tracked and signed in the CRM, with referral attribution fixed at entry so the numbers add up.

  • Weekly restocking sheets generated automatically, straight into purchasing
  • One data foundation: sales, warehouse and finance use the same definitions
  • Every step auditable, reproducible and traceable
CRM · open source

Education and admissions consulting

Admissions teams of 1–20 people, with customers living in WeChat and deals closing on trial classes and referrals. Daedalus CRM ships with exactly this vocabulary: student, school / year / major, trial attended, discussing with family. Referral attribution fixes “who referred this, whose number it counts toward” at the moment of entry. The referral radar shows who is actually bringing you people and who to ask next. Quick notes turn a WeChat conversation straight into a follow-up record.

  • Self-hosted, one command, your data stays yours
  • AI drafts but never writes; use DeepSeek or a local model
  • Not in education? Change the terms and fields in settings
GitHub · AGPL-3.0 ↗
Workflow · ChengguanEngineering practice

Engineering teams

Most of an engineering team's knowledge lives in individual heads: how to release, how to roll back, how to write a migration without losing data. Chengguan uses its rule engine and agents to turn that into reusable workflows — task handoffs, code collaboration, documentation kept in step, every step on the record. We work this way ourselves: the Daedalus CRM repository is a working template.

  • Three automatic gates before commit — type check, lint, unit tests — and they run on push
  • Migrations only add, never alter or drop; an old build still runs against the new database, so rollback touches no data
  • Concurrent writes merge per field and every write is logged — collaboration doesn't depend on shouting “I'm editing this” first
See how we work ↗
Not your situation?

Tell us your process first.

All three product lines grew out of real operations and share one data and process model. Walk us through yours and we'll tell you which one or two fit, and which steps should not use AI at all. The demo runs on your real data, not slideware.

Book a demo →
Try it yourself first

The application layer is open source.

No forms, no sales call. Three minutes to run it on your own machine — not a single row of data leaves your network