@zhixuan92/multi-model-agent-mcp 0.1.3 → 0.1.4
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +19 -3
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -54,22 +54,24 @@ Provider auth:
|
|
|
54
54
|
|
|
55
55
|
### Claude Code
|
|
56
56
|
|
|
57
|
-
One command — the client will spawn the server on demand:
|
|
57
|
+
One command — the client will spawn the server on demand. Use `-s user` so the server is available in **every** project on your machine, not just the directory where you ran the command:
|
|
58
58
|
|
|
59
59
|
```bash
|
|
60
|
-
claude mcp add multi-model-agent -- npx -y @zhixuan92/multi-model-agent-mcp serve
|
|
60
|
+
claude mcp add multi-model-agent -s user -- npx -y @zhixuan92/multi-model-agent-mcp serve
|
|
61
61
|
```
|
|
62
62
|
|
|
63
63
|
If your providers need environment variables:
|
|
64
64
|
|
|
65
65
|
```bash
|
|
66
|
-
claude mcp add multi-model-agent \
|
|
66
|
+
claude mcp add multi-model-agent -s user \
|
|
67
67
|
-e OPENAI_API_KEY=sk-... \
|
|
68
68
|
-e ANTHROPIC_API_KEY=sk-ant-... \
|
|
69
69
|
-e MINIMAX_API_KEY=... \
|
|
70
70
|
-- npx -y @zhixuan92/multi-model-agent-mcp serve
|
|
71
71
|
```
|
|
72
72
|
|
|
73
|
+
Without `-s user`, `claude mcp add` defaults to local scope and only registers the server in the current project.
|
|
74
|
+
|
|
73
75
|
### Claude Desktop
|
|
74
76
|
|
|
75
77
|
Add to `claude_desktop_config.json`:
|
|
@@ -92,6 +94,20 @@ Add to `claude_desktop_config.json`:
|
|
|
92
94
|
|
|
93
95
|
Restart your MCP client after changing config.
|
|
94
96
|
|
|
97
|
+
## Recommended: delegation rule for Claude Code
|
|
98
|
+
|
|
99
|
+
Claude Code's native `Task` / `Agent` subagents inherit your parent session's expensive model and eat its context window. We ship a drop-in rule file that teaches Claude Code **when** to delegate work through `delegate_tasks` instead — mechanical edits go to free providers, reasoning-tier work escalates only when needed, and independent tasks run in parallel.
|
|
100
|
+
|
|
101
|
+
Install globally:
|
|
102
|
+
|
|
103
|
+
```bash
|
|
104
|
+
mkdir -p ~/.claude/rules
|
|
105
|
+
curl -o ~/.claude/rules/multi-model-delegation.md \
|
|
106
|
+
https://raw.githubusercontent.com/zhixuan312/multi-model-agent/HEAD/docs/claude-code-delegation-rule.md
|
|
107
|
+
```
|
|
108
|
+
|
|
109
|
+
Restart Claude Code after installing. The full rule — judgment-vs-labor principle, decision procedure, provider routing table, dispatch shape, verification, escalation ladder — lives at [`docs/claude-code-delegation-rule.md`](https://github.com/zhixuan312/multi-model-agent/blob/HEAD/docs/claude-code-delegation-rule.md). Read that file before adapting it to your own provider names.
|
|
110
|
+
|
|
95
111
|
## The `delegate_tasks` tool
|
|
96
112
|
|
|
97
113
|
Accepts an array of tasks and runs them concurrently. Auto-routes each task by capability filter → quality tier filter → cheapest remaining provider, or pin a task to a specific provider.
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@zhixuan92/multi-model-agent-mcp",
|
|
3
|
-
"version": "0.1.
|
|
3
|
+
"version": "0.1.4",
|
|
4
4
|
"type": "module",
|
|
5
5
|
"license": "MIT",
|
|
6
6
|
"description": "MCP server for multi-model-agent. Exposes a delegate_tasks tool that routes work to Claude, Codex, or OpenAI-compatible sub-agents based on capability, quality tier, and cost.",
|