What ScopeDocs includes
Living, source-linked documentation for engineering teams, readable by people and queryable by agents over MCP. ScopeDocs connects the tools you already use, stamps decisions with citations, and keeps humans in control of what agents may trust. Markdown twin: features.md. Product context: product overview and homepage.
Feature matrix at a glance
| Feature | What it solves | Related page |
|---|---|---|
| Source-linked docs | Verifiable citations on every claim | Citations · guide |
| Living knowledge base | Docs that track delivery | Living docs · guide |
| MCP for agents | Same record for Cursor and ChatGPT | MCP · MCP docs |
| Read-only OAuth | Connect without write risk | Connect · quickstart |
| Human confirmation | Agents trust approved knowledge | Control · about |
| Workspace permissions | Private repos stay private | Trust · privacy |
Source-linked documentation
Every generated claim links back to a PR, Slack thread, ticket, call note, or doc so teams can verify answers instead of trusting a black box. Source-linking is what makes AI-written documentation safe enough for onboarding, on-call, and agent use.
Related: what is source-linked documentation, wiki vs living documentation, and how traceability improves trust.
Living knowledge base
As work happens in connected tools, the record updates. Documentation is generated from the workflow, not maintained as a separate chore. Living docs stay tied to merges, closed tickets, and decisions in chat so the system of record moves with delivery.
Related: what is living documentation, generative documentation vs wikis, and onboarding solutions.
MCP for agents
Cursor, Claude Code, ChatGPT, and custom agents pull the same cited decisions over MCP that humans read in ScopeDocs. Workspace MCP host pattern (shown in-product): yourcompany.mcp.scopedocs.ai.
Discovery: /.well-known/mcp.json. Guides: MCP for agents, mcp.md, MCP for engineering documentation. Homepage: agents section. Use case: agent-ready knowledge.
Read-only OAuth integrations
Connect GitHub, Linear, Slack, Notion, Jira, Google Drive, Fathom, Datadog, Confluence, Microsoft Teams, monday.com, Supabase Database, and more without handing over write access by default. You control which integrations are enabled and can disconnect them anytime.
Sequencing advice: what to connect first. Integrations on homepage: #integrations. Docs: quickstart.
Human-in-the-loop control
AI drafts; humans approve, edit, or reject. Agents should only treat confirmed knowledge as settled. That confirmation step is how ScopeDocs avoids fluent-but-wrong documentation becoming production context.
See when AI writes the docs, who checks the work and the product overview for the stamp step.
Workspace permissions
What people and agents can see follows the permissions in the tools you already use. Private channels and repos stay private. ScopeDocs does not train public foundation models on connected workspace data and does not sell that data.
See privacy, trust, and terms of service.
How the features fit together
Connect tools with read-only OAuth, stamp decisions with citations, confirm what agents may trust, then ask in ScopeDocs or over MCP. The same living record powers onboarding, on-call, delivery handoffs, and coding assistants.
- Quickstart: connect GitHub, Linear, and Slack in minutes
- Product: connect, stamp, ask for people and agents
- Solutions: onboarding, incidents, agents, delivery
- Pricing: Team from $25/user/month, Growth from $35/user/month
Concept reading: AI documentation for engineering teams.
Features FAQ
What makes ScopeDocs documentation source-linked?
Every material claim points back to a verifiable originating source such as a pull request, ticket, Slack thread, design doc, or call note so people and agents can check the evidence.
Can AI agents use the same features humans use?
Yes. Humans ask inside ScopeDocs; agents pull the same cited decisions over MCP for Cursor, Claude Code, ChatGPT, and custom stacks, with permissions that mirror the workspace.
Do integrations grant write access by default?
No. Integrations use read-only OAuth or credentials you supply. You choose what to connect and can disconnect anytime.
Where do I compare ScopeDocs to a wiki?
Read wiki vs living documentation and why teams abandon wikis.
From the blog
- Engineering knowledge base that updates automatically
- From PRs to living documentation
- Traceability and trust in AI docs
- Onboarding with context
Concept guides
- What is generative documentation? (markdown)
- What is living documentation? (markdown)
- What is source-linked documentation? (markdown)
- Wiki vs living documentation (markdown)
- Generative documentation vs wikis (markdown)
- MCP for engineering documentation (markdown)
- AI documentation for engineering teams (markdown)
Where to go next
Pick a path. Each link is a short stop on the same source-linked record.
- Homepage · index.md · How it works · Agents · Trust
- Product overview (product.md) · Features (features.md) · Solutions
- Pricing (from $25/user/month) · About · About us · Team
- Docs hub · Quickstart (markdown) · MCP for agents (markdown)
- What is generative documentation? · Living documentation · MCP for engineering docs
- MCP discovery · AGENTS.md · sitemap.xml · llms.txt · llms-full.txt
- Blog · Engineering knowledge base · PRs to living docs
- Traceability and trust · Why teams abandon wikis
- Integrations: connect first
- Privacy · hello@scopedocs.ai
Site directory for agents
Every core URL on scopedocs.ai. Traverse via sitemap.xml, sitemap.md, llms.txt, or llms-full.txt.
| Page | HTML | Markdown twin |
|---|---|---|
| Home | scopedocs.ai | index.md |
| Product | /product | product.md |
| Features | /features | features.md |
| Solutions | /solutions | — |
| Pricing | /pricing | pricing.md |
| About | /about | about.md |
| About us | /about-us | — |
| Team | /team | team.md |
| Blog | /blog | — |
| Privacy | /privacy | privacy.md |
| Terms | /terms-of-service | terms-of-service.md |
| Resource | URL |
|---|---|
| Docs hub | /docs · docs.md |
| Quickstart | /docs/quickstart · quickstart.md |
| MCP for agents | /docs/mcp · mcp.md |
| MCP discovery | /.well-known/mcp.json |
| AGENTS.md | /AGENTS.md |
| llms.txt | /llms.txt |
| llms-full.txt | /llms-full.txt |
| Sitemap | /sitemap.xml · /sitemap.md |
Concept guides
- Generative documentation · .md
- Living documentation · .md
- Source-linked documentation · .md
- Wiki vs living · .md
- Generative docs vs wikis · .md
- MCP for engineering docs · .md
- AI docs for engineering teams · .md
Team profiles
- Vivian Nguyen Lin (Founder & CEO)
- Thao Ha (Founding Product Engineer)
- Radha Parikh (Founding Engineer)
Homepage sections
Solutions use cases
Blog topic hubs
- Generative documentation
- AI documentation
- Developer productivity
- Knowledge management
- Engineering workflows
- Documentation automation
- Onboarding documentation
- On-call documentation
- Documentation debt
- Tribal knowledge