MCP

Crawls stay on this device until you sync one. The synced copy lives in your organization's secure cloud storage, and that copy is what MCP clients connect to. Nothing reaches an external client unless you push it there.

Silkra MCP is read-only. Clients can list synced sites, filter pages, read content, and find similar pages. They cannot crawl or change the local project. Cloud sync and MCP are paid.

Cloud sync

On any workspace, use the sidebar cloud button or search Manage MCP with ⌘P. Sync now uploads the latest complete crawl. That is the one-click turn-on.

Keep synced means later crawls upload on their own. It does not start an upload. Only a finished crawl is published. Re-sync after a refresh if clients should see the new version. Remove cloud copy deletes the synced version without touching the workspace on this device.

Any workspace, any client

Once cloud sync is on, a connected MCP client can query any workspace in the storage you can see. You do not add a server per site.

https://mcp.usesilkra.com/api/mcp

Sign-in happens on usesilkra.com. Approve access once per client. You may need to click Connect again in the client after the browser returns.

Team and Agency storage

Team and Agency plans group crawl storage into one shared pool. Anyone on the plan can query those synced workspaces from their own MCP client. They see what the team pushed, not a crawl that only exists on someone else's machine.

Claude

Paste the MCP URL as a custom connector. Connect Claude →

ChatGPT

Requires Developer Mode. Connect ChatGPT →

Codex

Add the remote server in config.toml, then log in. Connect Codex →

Cursor

Add the remote server in mcp.json, or let an agent do it. Connect Cursor →

Starter prompts

Paste these after the client is connected. Replace the site name or topic as needed.

Inventory

Using Silkra MCP, list my synced sites and summarize page counts for each.

Issue triage

Using Silkra, what are the highest-severity SEO issues on [site]? Group them by theme and suggest a fix order.

Content gaps

Search my synced Silkra crawl for pages about [topic]. Which URLs are strongest, which are thin, and what titles would you improve?

Cannibalization

Find pages on my Silkra crawl that compete for the same intent around [query]. List overlapping URLs and a consolidation recommendation.

Brief from evidence

Pull page content for the top matching Silkra pages about [topic], then draft a short content brief with target URL, angle, and internal links to reuse.

What clients can ask

Start with the site list, then a brief, then issues, filters, content, or similar pages.

Troubleshooting

Agents → Connections →