ai-based-api

Name: ai-based-api
Author: LambdaTest

335stars67forks

How to install ai-based-api?

lambdatest/agent-skills/ai-based-api

$npx -y skills add LambdaTest/agent-skills --skill ai-based-api

Installs into the current project.

Summary

Designs AI-powered API features, LLM tool/function definitions, MCP server tool schemas, natural language to API conversion, and agentic API workflows. Use whenever the user asks about "AI calling my API", "function calling schema", "tool definition for LLM", "MCP tools", "natura

Files · 1

View on GitHub

SKILL.md

1	# AI-Augmented API Skill
2
3	Design LLM tool definitions, agentic workflows, and natural language API interfaces.
4
5	---
6
7	## Anthropic Tool Use Definition
8
9	```json
10	{
11	"name": "search_products",
12	"description": "Search for products by keyword, category, or price range. Use when the user wants to find, browse, or compare products.",
13	"input_schema": {
14	"type": "object",
15	"properties": {
16	"query": {
17	"type": "string",
18	"description": "Search query keywords"
19	},
20	"category": {
21	"type": "string",
22	"enum": ["electronics", "clothing", "books", "home"],
23	"description": "Optional category filter"
24	},
25	"min_price": { "type": "number", "description": "Minimum price in USD" },
26	"max_price": { "type": "number", "description": "Maximum price in USD" },
27	"limit": { "type": "integer", "default": 10, "description": "Max results to return" }
28	},
29	"required": ["query"]
30	}
31	}
32	```
33
34	---
35
36	## OpenAI Function Calling Definition
37
38	```json
39	{
40	"type": "function",
41	"function": {
42	"name": "create_order",
43	"description": "Create a new order for a user. Use when the user wants to purchase a product. Always confirm product and quantity before calling.",
44	"parameters": {
45	"type": "object",
46	"properties": {
47	"product_id": { "type": "string", "description": "The product ID to order" },
48	"quantity": { "type": "integer", "minimum": 1, "description": "Quantity to order" },
49	"shipping_address": {
50	"type": "object",
51	"properties": {
52	"street": { "type": "string" },
53	"city": { "type": "string" },
54	"country": { "type": "string" }
55	},
56	"required": ["street", "city", "country"]
57	}
58	},
59	"required": ["product_id", "quantity", "shipping_address"]
60	}
61	}
62	}
63	```
64
65	---
66
67	## MCP (Model Context Protocol) Tool Schema
68
69	```json
70	{
71	"name": "get_build_status",
72	"description": "Get the status of a HyperExecute test job. Use when the user asks about test results, job status, or CI build outcomes.",
73	"inputSchema": {
74	"type": "object",
75	"properties": {
76	"job_id": { "type": "string", "description": "The HyperExecute job ID" }
77	},
78	"required": ["job_id"]
79	}
80	}
81	```
82
83	> 🔗 Real-World Integration — TestMu AI HyperExecute
84	> Build MCP tools that let AI agents query and control test jobs via the HyperExecute API.
85	> Docs: https://www.testmuai.com/support/api-doc/?key=hyperexecute
86
87	---
88
89	## Tool Design Principles
90
91	1. One tool = one action: Don't combine search + filter + sort into one tool. Split them.
92	2. Description drives routing: The LLM picks tools from descriptions — be specific and include trigger phrases.
93	3. Required vs optional: Only mark fields `required` if the API truly needs them.
94	4. Enum for constrained values: Use `enum` instead of `string` for fixed-choice fields.
95	5. Idempotent where possible: Prefer read tools over write tools for exploration.
96	6. Confirm before destructive actions: Description should say "Always confirm with the user before calling."
97
98	---
99
100	## Agentic Workflow Example
101
102	```
103	User: "Get me the status of my last 3 test builds"
104
105	Agent plan:
106	1. call list_jobs(limit=3, sort="created_at:desc")
107	→ returns [{id: "job_1", status: "passed"}, {id: "job_2", status: "failed"}, ...]
108	2. call get_job_details(job_id="job_2") // dig into the failed one
109	→ returns task breakdown, error logs
110	3. Synthesize: "Your last 3 builds: job_1 passed, job_2 failed (2 of 15 tasks failed on Chrome/Win10), job_3 passed."
111	```
112
113	---
114
115	## Natural Language → API Mapping Table
116
117	Build this mapping for any domain:
118
119	\| Natural language intent \| API call \|
120	\|------------------------\|---------\|
121	\| "Find hotels in Paris" \| `GET /hotels/search?location=Paris` \|
122	\| "Book a room for 2 nights" \| `POST /bookings` \|
123	\| "Cancel my reservation" \| `POST /bookings/{id}/cancel` \|
124	\| "Show my past orders" \| `GET /orders?user=me&sort=date:desc` \|
125	\| "Is the API working?" \| `GET /health/ready` \|
126
127	---
128
129	## API-as-Plugin (OpenAPI → GPT Plugin / Tool)
130
131	Minimal `ai-plugin.json`:
132	```json
133	{
134	"schema_version": "v1",
135	"name_for_human": "My API",
136	"name_for_model": "my_api",
137	"description_for

People who liked this also viewed

6 picks

Type

ai-based-api

335stars67forks

View on GitHub

How to install ai-based-api?

lambdatest/agent-skills/ai-based-api

$npx -y skills add LambdaTest/agent-skills --skill ai-based-api

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# AI-Augmented API Skill
2
3	Design LLM tool definitions, agentic workflows, and natural language API interfaces.
4
5	---
6
7	## Anthropic Tool Use Definition
8
9	```json
10	{
11	"name": "search_products",
12	"description": "Search for products by keyword, category, or price range. Use when the user wants to find, browse, or compare products.",
13	"input_schema": {
14	"type": "object",
15	"properties": {
16	"query": {
17	"type": "string",
18	"description": "Search query keywords"
19	},
20	"category": {
21	"type": "string",
22	"enum": ["electronics", "clothing", "books", "home"],
23	"description": "Optional category filter"
24	},
25	"min_price": { "type": "number", "description": "Minimum price in USD" },
26	"max_price": { "type": "number", "description": "Maximum price in USD" },
27	"limit": { "type": "integer", "default": 10, "description": "Max results to return" }
28	},
29	"required": ["query"]
30	}
31	}
32	```
33
34	---
35
36	## OpenAI Function Calling Definition
37
38	```json
39	{
40	"type": "function",
41	"function": {
42	"name": "create_order",
43	"description": "Create a new order for a user. Use when the user wants to purchase a product. Always confirm product and quantity before calling.",
44	"parameters": {
45	"type": "object",
46	"properties": {
47	"product_id": { "type": "string", "description": "The product ID to order" },
48	"quantity": { "type": "integer", "minimum": 1, "description": "Quantity to order" },
49	"shipping_address": {
50	"type": "object",
51	"properties": {
52	"street": { "type": "string" },
53	"city": { "type": "string" },
54	"country": { "type": "string" }
55	},
56	"required": ["street", "city", "country"]
57	}
58	},
59	"required": ["product_id", "quantity", "shipping_address"]
60	}
61	}
62	}
63	```
64
65	---
66
67	## MCP (Model Context Protocol) Tool Schema
68
69	```json
70	{
71	"name": "get_build_status",
72	"description": "Get the status of a HyperExecute test job. Use when the user asks about test results, job status, or CI build outcomes.",
73	"inputSchema": {
74	"type": "object",
75	"properties": {
76	"job_id": { "type": "string", "description": "The HyperExecute job ID" }
77	},
78	"required": ["job_id"]
79	}
80	}
81	```
82
83	> 🔗 Real-World Integration — TestMu AI HyperExecute
84	> Build MCP tools that let AI agents query and control test jobs via the HyperExecute API.
85	> Docs: https://www.testmuai.com/support/api-doc/?key=hyperexecute
86
87	---
88
89	## Tool Design Principles
90
91	1. One tool = one action: Don't combine search + filter + sort into one tool. Split them.
92	2. Description drives routing: The LLM picks tools from descriptions — be specific and include trigger phrases.
93	3. Required vs optional: Only mark fields `required` if the API truly needs them.
94	4. Enum for constrained values: Use `enum` instead of `string` for fixed-choice fields.
95	5. Idempotent where possible: Prefer read tools over write tools for exploration.
96	6. Confirm before destructive actions: Description should say "Always confirm with the user before calling."
97
98	---
99
100	## Agentic Workflow Example
101
102	```
103	User: "Get me the status of my last 3 test builds"
104
105	Agent plan:
106	1. call list_jobs(limit=3, sort="created_at:desc")
107	→ returns [{id: "job_1", status: "passed"}, {id: "job_2", status: "failed"}, ...]
108	2. call get_job_details(job_id="job_2") // dig into the failed one
109	→ returns task breakdown, error logs
110	3. Synthesize: "Your last 3 builds: job_1 passed, job_2 failed (2 of 15 tasks failed on Chrome/Win10), job_3 passed."
111	```
112
113	---
114
115	## Natural Language → API Mapping Table
116
117	Build this mapping for any domain:
118
119	\| Natural language intent \| API call \|
120	\|------------------------\|---------\|
121	\| "Find hotels in Paris" \| `GET /hotels/search?location=Paris` \|
122	\| "Book a room for 2 nights" \| `POST /bookings` \|
123	\| "Cancel my reservation" \| `POST /bookings/{id}/cancel` \|
124	\| "Show my past orders" \| `GET /orders?user=me&sort=date:desc` \|
125	\| "Is the API working?" \| `GET /health/ready` \|
126
127	---
128
129	## API-as-Plugin (OpenAPI → GPT Plugin / Tool)
130
131	Minimal `ai-plugin.json`:
132	```json
133	{
134	"schema_version": "v1",
135	"name_for_human": "My API",
136	"name_for_model": "my_api",
137	"description_for

People who liked this also viewed

6 picks

Type