mcp-client

Name: mcp-client
Author: walidboulanouar

82stars17forks

How to install mcp-client?

walidboulanouar/ay-skills/mcp-client

$npx -y skills add walidboulanouar/Ay-Skills --skill mcp-client

Installs into the current project.

Summary

Universal MCP client for connecting to any MCP server with progressive disclosure. Wraps MCP servers as skills to avoid context window bloat from tool definitions. Use when interacting with external MCP servers (Zapier, Sequential Thinking, GitHub, filesystem, etc.), listing avai

Files · 1

View on GitHub

SKILL.md

1	# Universal MCP Client
2
3	Connect to any MCP server with progressive disclosure - load tool schemas on-demand instead of dumping thousands of tokens into context upfront.
4
5	## Skill Location
6
7	This skill is located at: `.claude/skills/mcp-client/`
8
9	Script path: `.claude/skills/mcp-client/scripts/mcp_client.py`
10
11	## Configuration
12
13	The script looks for config in this order:
14	1. `MCP_CONFIG_PATH` env var (custom path)
15	2. `references/mcp-config.json` (this skill's config - recommended)
16	3. `.mcp.json` in project root
17	4. `~/.claude.json`
18
19	Your config file: `.claude/skills/mcp-client/references/mcp-config.json`
20
21	Edit this file to add your API keys. The example file (`example-mcp-config.json`) is kept as a reference template.
22
23	Security note:
24	- Never commit real API keys. This repo uses a placeholder `${AY_PROPOSALS_API_KEY}` for the AY Proposals server.
25	- Option 1 (recommended): Create a private config and point the script to it: `export MCP_CONFIG_PATH=~/.claude/skills/mcp-client/references/mcp-config.local.json` (same structure, with your real token). Keep that file out of git.
26	- Option 2: Export your key in the shell: `export AY_PROPOSALS_API_KEY=...` and ensure your runtime interpolates env vars. If not, use Option 1.
27
28	If the user hasn't provided their Zapier API key yet, ask them for it.
29
30	## Running Commands
31
32	All commands use the script at `.claude/skills/mcp-client/scripts/mcp_client.py`:
33
34	```bash
35	# List configured servers
36	python .claude/skills/mcp-client/scripts/mcp_client.py servers
37
38	# List tools from a server
39	python .claude/skills/mcp-client/scripts/mcp_client.py tools <server_name>
40
41	# Call a tool
42	python .claude/skills/mcp-client/scripts/mcp_client.py call <server> <tool> '{"arg": "value"}'
43	```
44
45	## Workflow
46
47	1. Check config exists - Run `servers` command. If error, create `.mcp.json`
48	2. List servers - See what MCP servers are configured
49	3. List tools - Get tool schemas from a specific server
50	4. Call tool - Execute a tool with arguments
51
52	## Commands Reference
53
54	\| Command \| Description \|
55	\|---------\|-------------\|
56	\| `servers` \| List all configured MCP servers \|
57	\| `tools <server>` \| List tools with full parameter schemas \|
58	\| `call <server> <tool> '<json>'` \| Execute a tool with arguments \|
59
60	## Example: Zapier
61
62	```bash
63	# 1. List servers to confirm Zapier is configured
64	python .claude/skills/mcp-client/scripts/mcp_client.py servers
65
66	# 2. List Zapier tools
67	python .claude/skills/mcp-client/scripts/mcp_client.py tools zapier
68
69	# 3. Call a Zapier tool
70	python .claude/skills/mcp-client/scripts/mcp_client.py call zapier <tool_name> '{"param": "value"}'
71	```
72
73	## Example: Sequential Thinking
74
75	```bash
76	# 1. List tools
77	python .claude/skills/mcp-client/scripts/mcp_client.py tools sequential-thinking
78
79	# 2. Use sequential thinking
80	python .claude/skills/mcp-client/scripts/mcp_client.py call sequential-thinking sequentialthinking '{"thought": "Breaking down the problem...", "thoughtNumber": 1, "totalThoughts": 5, "nextThoughtNeeded": true}'
81	```
82
83	## Config Format
84
85	Config file format (`references/mcp-config.json`):
86
87	```json
88	{
89	"mcpServers": {
90	"zapier": {
91	"url": "https://mcp.zapier.com/api/v1/connect",
92	"api_key": "your-api-key"
93	},
94	"sequential-thinking": {
95	"command": "npx",
96	"args": ["-y", "@modelcontextprotocol/server-sequential-thinking"]
97	}
98	}
99	}
100	```
101
102	Transport detection:
103	- `url` + `api_key` → FastMCP with Bearer auth (Zapier)
104	- `command` + `args` → stdio (local servers like sequential-thinking)
105	- `url` ending in `/sse` → SSE transport
106	- `url` ending in `/mcp` → Streamable HTTP
107
108	## Error Handling
109
110	Errors return JSON:
111	```json
112	{"error": "message", "type": "configuration\|validation\|connection"}
113	```
114
115	- `configuration` - Config file not found. Create `.mcp.json`
116	- `validation` - Invalid server or tool name
117	- `connection` - Failed to connect to server
118
119	## Dependencies
120
121	```bash
122	pip install mcp fastmcp
123	```
124
125	## References
126
127	- `references/example-mcp-config.json` - Template config file
128	- `references/mcp-servers.md` - Common server configurations
129	- `references/python-mcp-sdk.md` - Python SDK documentation

People who liked this also viewed

6 picks

Type

mcp-client

82stars17forks

View on GitHub

How to install mcp-client?

walidboulanouar/ay-skills/mcp-client

$npx -y skills add walidboulanouar/Ay-Skills --skill mcp-client

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Universal MCP Client
2
3	Connect to any MCP server with progressive disclosure - load tool schemas on-demand instead of dumping thousands of tokens into context upfront.
4
5	## Skill Location
6
7	This skill is located at: `.claude/skills/mcp-client/`
8
9	Script path: `.claude/skills/mcp-client/scripts/mcp_client.py`
10
11	## Configuration
12
13	The script looks for config in this order:
14	1. `MCP_CONFIG_PATH` env var (custom path)
15	2. `references/mcp-config.json` (this skill's config - recommended)
16	3. `.mcp.json` in project root
17	4. `~/.claude.json`
18
19	Your config file: `.claude/skills/mcp-client/references/mcp-config.json`
20
21	Edit this file to add your API keys. The example file (`example-mcp-config.json`) is kept as a reference template.
22
23	Security note:
24	- Never commit real API keys. This repo uses a placeholder `${AY_PROPOSALS_API_KEY}` for the AY Proposals server.
25	- Option 1 (recommended): Create a private config and point the script to it: `export MCP_CONFIG_PATH=~/.claude/skills/mcp-client/references/mcp-config.local.json` (same structure, with your real token). Keep that file out of git.
26	- Option 2: Export your key in the shell: `export AY_PROPOSALS_API_KEY=...` and ensure your runtime interpolates env vars. If not, use Option 1.
27
28	If the user hasn't provided their Zapier API key yet, ask them for it.
29
30	## Running Commands
31
32	All commands use the script at `.claude/skills/mcp-client/scripts/mcp_client.py`:
33
34	```bash
35	# List configured servers
36	python .claude/skills/mcp-client/scripts/mcp_client.py servers
37
38	# List tools from a server
39	python .claude/skills/mcp-client/scripts/mcp_client.py tools <server_name>
40
41	# Call a tool
42	python .claude/skills/mcp-client/scripts/mcp_client.py call <server> <tool> '{"arg": "value"}'
43	```
44
45	## Workflow
46
47	1. Check config exists - Run `servers` command. If error, create `.mcp.json`
48	2. List servers - See what MCP servers are configured
49	3. List tools - Get tool schemas from a specific server
50	4. Call tool - Execute a tool with arguments
51
52	## Commands Reference
53
54	\| Command \| Description \|
55	\|---------\|-------------\|
56	\| `servers` \| List all configured MCP servers \|
57	\| `tools <server>` \| List tools with full parameter schemas \|
58	\| `call <server> <tool> '<json>'` \| Execute a tool with arguments \|
59
60	## Example: Zapier
61
62	```bash
63	# 1. List servers to confirm Zapier is configured
64	python .claude/skills/mcp-client/scripts/mcp_client.py servers
65
66	# 2. List Zapier tools
67	python .claude/skills/mcp-client/scripts/mcp_client.py tools zapier
68
69	# 3. Call a Zapier tool
70	python .claude/skills/mcp-client/scripts/mcp_client.py call zapier <tool_name> '{"param": "value"}'
71	```
72
73	## Example: Sequential Thinking
74
75	```bash
76	# 1. List tools
77	python .claude/skills/mcp-client/scripts/mcp_client.py tools sequential-thinking
78
79	# 2. Use sequential thinking
80	python .claude/skills/mcp-client/scripts/mcp_client.py call sequential-thinking sequentialthinking '{"thought": "Breaking down the problem...", "thoughtNumber": 1, "totalThoughts": 5, "nextThoughtNeeded": true}'
81	```
82
83	## Config Format
84
85	Config file format (`references/mcp-config.json`):
86
87	```json
88	{
89	"mcpServers": {
90	"zapier": {
91	"url": "https://mcp.zapier.com/api/v1/connect",
92	"api_key": "your-api-key"
93	},
94	"sequential-thinking": {
95	"command": "npx",
96	"args": ["-y", "@modelcontextprotocol/server-sequential-thinking"]
97	}
98	}
99	}
100	```
101
102	Transport detection:
103	- `url` + `api_key` → FastMCP with Bearer auth (Zapier)
104	- `command` + `args` → stdio (local servers like sequential-thinking)
105	- `url` ending in `/sse` → SSE transport
106	- `url` ending in `/mcp` → Streamable HTTP
107
108	## Error Handling
109
110	Errors return JSON:
111	```json
112	{"error": "message", "type": "configuration\|validation\|connection"}
113	```
114
115	- `configuration` - Config file not found. Create `.mcp.json`
116	- `validation` - Invalid server or tool name
117	- `connection` - Failed to connect to server
118
119	## Dependencies
120
121	```bash
122	pip install mcp fastmcp
123	```
124
125	## References
126
127	- `references/example-mcp-config.json` - Template config file
128	- `references/mcp-servers.md` - Common server configurations
129	- `references/python-mcp-sdk.md` - Python SDK documentation

People who liked this also viewed

6 picks

Type