Issue Station
Sign In
Tools hereToolbox

openclaw - 💡(How to fix) Fix [Feature]: Expose MCP capabilities through gateway RPC methods

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…

OpenClaw currently exposes MCP mainly through the local HTTP /mcp endpoint, but not through gateway RPC methods such as mcp.*. I would like to understand whether this separation is intentional and whether MCP-related runtime capabilities could also be exposed through the gateway RPC system.

Root Cause

OpenClaw currently exposes MCP mainly through the local HTTP /mcp endpoint, but not through gateway RPC methods such as mcp.*. I would like to understand whether this separation is intentional and whether MCP-related runtime capabilities could also be exposed through the gateway RPC system.

RAW_BUFFERClick to expand / collapse

Summary

OpenClaw currently exposes MCP mainly through the local HTTP /mcp endpoint, but not through gateway RPC methods such as mcp.*. I would like to understand whether this separation is intentional and whether MCP-related runtime capabilities could also be exposed through the gateway RPC system.

Problem to solve

While exploring the OpenClaw architecture, I noticed that gateway RPC currently exposes methods like:

  • skills.*
  • tools.*
  • sessions.*
  • config.*
  • agent.*

However, MCP functionality appears to exist separately under:

  • mcp-http.handlers.ts
  • mcp-http.request.ts
  • mcp-http.loopback-runtime.ts

and is exposed through the HTTP /mcp endpoint instead of gateway RPC methods.

This makes it difficult to:

  • introspect MCP state from the UI/runtime layer
  • manage MCP servers through the existing gateway RPC system
  • build custom UI/runtime integrations around MCP
  • keep architectural consistency between tools/skills/runtime and MCP capabilities

Would it make sense to expose MCP-related capabilities through gateway RPC methods as well? For example:

  • mcp.list
  • mcp.servers
  • mcp.tools
  • mcp.reload

I would also like to understand whether the current separation is an intentional architectural boundary.

Proposed solution

none

Alternatives considered

No response

Impact

none

Evidence/examples

No response

Additional information

No response

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…
#retriever error#indexing error#inference speed#output truncation#response parsing

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING

openclaw - 💡(How to fix) Fix [Feature]: Expose MCP capabilities through gateway RPC methods