Concepts
The core vocabulary behind Workspace Intelligence and software system understanding.
Workspace Intelligence is the shared, evidence-backed understanding layer for software systems.
It is not a coding agent, an IDE, a repository indexer, a RAG database, or a chat window. Those tools may consume Workspace Intelligence, but they are not the layer itself.
For the canonical definition, start with What Is Workspace Intelligence?.
The problem
AI tools often receive more context without receiving better understanding.
They can read files, embeddings, snippets, issues, and generated summaries. But production software systems are larger than code text:
- multiple projects,
- generated and adopted applications,
- runtime commands,
- ownership,
- policies,
- CI evidence,
- release gates,
- agent instructions,
- stale facts,
- operational decisions.
Calling all of this "context" hides the real problem. Different layers serve different jobs.
| Layer | Purpose | Typical output |
|---|---|---|
| Skills | Teach procedures | Playbooks and reusable workflows |
| RAG | Retrieve knowledge | Retrieved chunks and citations |
| Memory | Remember user or session facts | Preferences and prior interactions |
| Repository Intelligence | Explain codebase structure | Files, symbols, imports, code graph |
| Workspace Intelligence | Understand the software system | Model, graph, evidence, context, impact, verify |
For the deeper distinction, read Context vs Understanding.
Definition
Workspace Intelligence is the deterministic layer that turns a software workspace into a shared operating model for humans and tools.
That model answers:
- What exists?
- How is it related?
- What changed?
- What is affected?
- What evidence supports the claim?
- What should humans, CI, IDEs, and AI agents trust right now?
Primary promise
Workspai turns repositories, projects, dependencies, rules, changes, and evidence into shared understanding for developers, CI, IDEs, and AI agents.
Core loop
The loop is intentionally simple:
Each step should be backed by contracts or generated evidence, not marketing copy.
Why evidence matters
Evidence is the source of understanding; documentation is an output, not the authority.
Docs can explain the architecture. The CLI, contracts, reports, and verification artifacts decide what is actually true for a workspace.