connect-apps-plugin
Composio Claude Agents SDK - 全局外部应用连接插件
Install with Tessl CLI
npx tessl i github:Lingjie-chen/MT5 --skill connect-apps-plugin58
Quality
36%
Does it follow best practices?
Impact
100%
3.33xAverage score across 3 eval scenarios
Optimize this skill with Tessl
npx tessl skill review --optimize ./.trae/skills/connect-apps-plugin/SKILL.mdEvaluation results
100%
88%Developer Automation Bot
Core Composio agent setup
composio_claude_agent_sdk import
0%
100%
claude_agent_sdk imports
0%
100%
Composio with provider
0%
100%
Session creation
0%
100%
Tools retrieval
0%
100%
MCP server name
0%
100%
mcp_servers in options
0%
100%
permission_mode set
0%
100%
ClaudeSDKClient context manager
0%
100%
client.query() call
0%
100%
receive_response loop
0%
100%
API key from environment
100%
100%
Without context: $1.0914 · 2m 53s · 34 turns · 211 in / 6,493 out tokens
With context: $0.4683 · 1m 46s · 23 turns · 442 in / 3,984 out tokens
100%
38%Multi-User External App Authorization Module
User auth and connection management
manage_connections disabled
50%
100%
session.authorize() usage
100%
100%
HTTPS callback URL
100%
100%
redirect_url accessed
100%
100%
wait_for_connection called
100%
100%
ConnectionError handling
50%
100%
list_connections called
50%
100%
delete_connection called
50%
100%
Multi-account label
20%
100%
ClaudeAgentSDKProvider used
0%
100%
API key from env var
66%
100%
Without context: $2.4674 · 9m 32s · 75 turns · 481 in / 21,676 out tokens
With context: $1.6074 · 6m 54s · 46 turns · 584 in / 20,716 out tokens
100%
84%Trading Alerts Agent
Toolkit filtering and permission config
Toolkit filtering
50%
100%
MCP server version
0%
100%
MCP server name
0%
100%
Confirmation permission mode
0%
100%
ClaudeSDKClient context manager
0%
100%
client.query() call
0%
100%
receive_response loop
0%
100%
ConnectionError handling
0%
100%
list_connections called
0%
100%
API key from env var
100%
100%
ClaudeAgentSDKProvider used
0%
100%
Without context: $2.7636 · 6m 56s · 38 turns · 235 in / 14,356 out tokens
With context: $0.4509 · 5m 28s · 21 turns · 392 in / 4,235 out tokens
- Evaluated
- Agent
- Claude Code
- Model
- Claude Sonnet 4.6
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.