Docs

Connect an agent to your feed

SKULayer ships an MCP server, so Claude and other MCP clients can read your feeds, audits, and items, and preview rule changes against real data. Ask "what's wrong with my feed?" and get the audit, not a guess.

What can a connected agent do?

Version 1 is read plus preview, deliberately. An agent can see everything and test anything, but applying a change stays in the app, behind the same preview and confirmation every rule goes through. Your feed cannot be modified from outside.

list_feeds

Every feed in the workspace with schedule, channels, item count, and the latest run summary.

get_feed

Full detail for one feed, including its rule stack size.

list_rules

The rule stack in execution order, each with a plain-English summary and its JSON definition.

get_audit_findings

Audit findings for the latest run, grouped by check, with why each one costs money and how to fix it.

search_items

Search or page through delivered items (what channels receive), or the base items before rules.

preview_rule

Test an unsaved rule against real items: match counts, field changes, warnings, and audit deltas. Writes nothing.

How do I connect?

  1. Create an API token. In the app, go to Account and create a token under API tokens (workspace owners only). Copy it when it's shown; it appears exactly once.
  2. Add the server to your client. The endpoint is https://skulayer.com/mcp (Streamable HTTP). For Claude Code: 
    claude mcp add --transport http skulayer https://skulayer.com/mcp \
  --header "Authorization: Bearer skl_your_token_here"
    Other MCP clients: point them at the URL and send the token as an Authorization: Bearer header.
  3. Ask about your feed. Start with something like "list my feeds and summarize the audit findings on the worst one", or "preview a rule that excludes items under $5 and tell me what it would change."

How is access controlled?

Tokens are scoped to one workspace and stored as hashes; we cannot read them back, and neither can anyone else. Every tool call is filtered to the token's workspace. Revoke a token from the Account page at any time and it stops working immediately. The server is read-only plus preview in this version, so the worst a leaked token can do is read the product data you already publish to ad platforms. Revoke it anyway.