cate.mcp: install endpoint into agents + default Dashboard - #18
Merged
Conversation
Install into agents (Endpoint screen): one click writes the aggregated /mcp endpoint into each agent's workspace-local config, each in that agent's own shape and preserving other keys. Cate Agent (.pi/mcp.json), Claude Code (.mcp.json), Cursor (.cursor/mcp.json), OpenCode (opencode.json), Codex (.codex/config.toml). Antigravity and PI are listed but disabled with the reason. Stale installs (the port reminted each session) surface an Update action. Default Dashboard: the panel opens on server health plus a live feed of tool calls through the endpoint. A bounded ActivityLog is recorded at the aggregator and served on its own /api/activity route, kept off /api/state so recording never churns the state serial. Bumps to v1.3.0.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Two additions to the cate.mcp extension (v1.3.0).
Install into agents (Endpoint screen)
One click writes the aggregated
/mcpendpoint into each agent's workspace-local config, each in that agent's own shape, preserving other keys:.pi/mcp.json(read by thepi-mcp-adapterpackage).mcp.json.cursor/mcp.jsonopencode.json.codex/config.tomlAntigravity and PI are listed but disabled with the reason (no workspace-local HTTP MCP config today). The endpoint's port/token are reminted each session, so an install is a snapshot: a stale entry is flagged with an Update action, and the endpoint only serves while the panel is open.
Default Dashboard
The panel now opens here instead of auto-selecting the first server: an endpoint strip, a server health grid, and a live Activity feed of tool calls through the endpoint. A bounded
ActivityLogring is recorded at the aggregator and served on its own/api/activityroute, kept off/api/stateso recording never churns the state serial.Verified
typecheck clean, 113 tests (including a real end-to-end assertion that a tool call through
/mcplands in/api/activity), build clean, sync-kit clean.Note: the Cate Agent target pairs with the main repo change that auto-registers
pi-mcp-adapter(0-AI-UG/cate#434).