fetching-library-docs

Name: fetching-library-docs
Author: mjunaidca

22installs11forks

How to install fetching-library-docs?

mjunaidca/mjs-agent-skills/fetching-library-docs

$npx -y skills add mjunaidca/mjs-agent-skills --skill fetching-library-docs

Installs into the current project.

Summary

Token-efficient library API documentation fetcher using Context7 MCP with 77% token savings. Fetches code examples, API references, and usage patterns for published libraries (React, Next.js, Prisma, etc). Use when users ask "how do I use X library", need code examples, want API

Files · 1

View on GitHub

SKILL.md

1	# Context7 Efficient Documentation Fetcher
2
3	Fetch library documentation with automatic 77% token reduction via shell pipeline.
4
5	## Quick Start
6
7	Always use the token-efficient shell pipeline:
8
9	```bash
10	# Automatic library resolution + filtering
11	bash scripts/fetch-docs.sh --library <library-name> --topic <topic>
12
13	# Examples:
14	bash scripts/fetch-docs.sh --library react --topic useState
15	bash scripts/fetch-docs.sh --library nextjs --topic routing
16	bash scripts/fetch-docs.sh --library prisma --topic queries
17	```
18
19	Result: Returns ~205 tokens instead of ~934 tokens (77% savings).
20
21	## Standard Workflow
22
23	For any documentation request, follow this workflow:
24
25	### 1. Identify Library and Topic
26
27	Extract from user query:
28	- Library: React, Next.js, Prisma, Express, etc.
29	- Topic: Specific feature (hooks, routing, queries, etc.)
30
31	### 2. Fetch with Shell Pipeline
32
33	```bash
34	bash scripts/fetch-docs.sh --library <library> --topic <topic> --verbose
35	```
36
37	The `--verbose` flag shows token savings statistics.
38
39	### 3. Use Filtered Output
40
41	The script automatically:
42	- Fetches full documentation (934 tokens, stays in subprocess)
43	- Filters to code examples + API signatures + key notes
44	- Returns only essential content (205 tokens to Claude)
45
46	## Parameters
47
48	### Basic Usage
49
50	```bash
51	bash scripts/fetch-docs.sh [OPTIONS]
52	```
53
54	Required (pick one):
55	- `--library <name>` - Library name (e.g., "react", "nextjs")
56	- `--library-id <id>` - Direct Context7 ID (faster, skips resolution)
57
58	Optional:
59	- `--topic <topic>` - Specific feature to focus on
60	- `--mode <code\|info>` - code for examples (default), info for concepts
61	- `--page <1-10>` - Pagination for more results
62	- `--verbose` - Show token savings statistics
63
64	### Mode Selection
65
66	Code Mode (default): Returns code examples + API signatures
67	```bash
68	--mode code
69	```
70
71	Info Mode: Returns conceptual explanations + fewer examples
72	```bash
73	--mode info
74	```
75
76	## Common Library IDs
77
78	Use `--library-id` for faster lookup (skips resolution):
79
80	```bash
81	React: /reactjs/react.dev
82	Next.js: /vercel/next.js
83	Express: /expressjs/express
84	Prisma: /prisma/docs
85	MongoDB: /mongodb/docs
86	Fastify: /fastify/fastify
87	NestJS: /nestjs/docs
88	Vue.js: /vuejs/docs
89	Svelte: /sveltejs/site
90	```
91
92	## Workflow Patterns
93
94	### Pattern 1: Quick Code Examples
95
96	User asks: "Show me React useState examples"
97
98	```bash
99	bash scripts/fetch-docs.sh --library react --topic useState --verbose
100	```
101
102	Returns: 5 code examples + API signatures + notes (~205 tokens)
103
104	### Pattern 2: Learning New Library
105
106	User asks: "How do I get started with Prisma?"
107
108	```bash
109	# Step 1: Get overview
110	bash scripts/fetch-docs.sh --library prisma --topic "getting started" --mode info
111
112	# Step 2: Get code examples
113	bash scripts/fetch-docs.sh --library prisma --topic queries --mode code
114	```
115
116	### Pattern 3: Specific Feature Lookup
117
118	User asks: "How does Next.js routing work?"
119
120	```bash
121	bash scripts/fetch-docs.sh --library-id /vercel/next.js --topic routing
122	```
123
124	Using `--library-id` is faster when you know the exact ID.
125
126	### Pattern 4: Deep Exploration
127
128	User needs comprehensive information:
129
130	```bash
131	# Page 1: Basic examples
132	bash scripts/fetch-docs.sh --library react --topic hooks --page 1
133
134	# Page 2: Advanced patterns
135	bash scripts/fetch-docs.sh --library react --topic hooks --page 2
136	```
137
138	## Token Efficiency
139
140	How it works:
141
142	1. `fetch-docs.sh` calls `fetch-raw.sh` (which uses `mcp-client.py`)
143	2. Full response (934 tokens) stays in subprocess memory
144	3. Shell filters (awk/grep/sed) extract essentials (0 LLM tokens used)
145	4. Returns filtered output (205 tokens) to Claude
146
147	Savings:
148	- Direct MCP: 934 tokens per query
149	- This approach: 205 tokens per query
150	- 77% reduction
151
152	Do NOT use `mcp-client.py` directly - it bypasses filtering and wastes tokens.
153
154	## Advanced: Library Resolution
155
156	If library name fails, try variations:
157
158	```bash
159	# Try different formats
160	--library "next.js" # with dot
161	--library "nextjs" # without dot
162	--library "next" # short form
163
164	# Or search manually
165	bash scripts/fetch-docs.sh --library "your-library" --verbose
166	# Check output for suggested library IDs
167	```
168
169	## Verification
170
171	Run: `python3 scripts/verify.py`
172
173	Expected: `✓ fetch-docs.sh ready`
174
175	## If Verification Fails
176
177	1. Run diagnostic: `ls -la scripts/fetch-docs.sh`
178	2. Check: Script exists and is executable
179	3. Fix: `chmod +x scripts/fetch-docs.sh`
180	4. Stop and report if still failing - do not proceed wi

People who liked this also viewed

6 picks

Type

fetching-library-docs

22installs11forks

View on GitHub

How to install fetching-library-docs?

mjunaidca/mjs-agent-skills/fetching-library-docs

$npx -y skills add mjunaidca/mjs-agent-skills --skill fetching-library-docs

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Context7 Efficient Documentation Fetcher
2
3	Fetch library documentation with automatic 77% token reduction via shell pipeline.
4
5	## Quick Start
6
7	Always use the token-efficient shell pipeline:
8
9	```bash
10	# Automatic library resolution + filtering
11	bash scripts/fetch-docs.sh --library <library-name> --topic <topic>
12
13	# Examples:
14	bash scripts/fetch-docs.sh --library react --topic useState
15	bash scripts/fetch-docs.sh --library nextjs --topic routing
16	bash scripts/fetch-docs.sh --library prisma --topic queries
17	```
18
19	Result: Returns ~205 tokens instead of ~934 tokens (77% savings).
20
21	## Standard Workflow
22
23	For any documentation request, follow this workflow:
24
25	### 1. Identify Library and Topic
26
27	Extract from user query:
28	- Library: React, Next.js, Prisma, Express, etc.
29	- Topic: Specific feature (hooks, routing, queries, etc.)
30
31	### 2. Fetch with Shell Pipeline
32
33	```bash
34	bash scripts/fetch-docs.sh --library <library> --topic <topic> --verbose
35	```
36
37	The `--verbose` flag shows token savings statistics.
38
39	### 3. Use Filtered Output
40
41	The script automatically:
42	- Fetches full documentation (934 tokens, stays in subprocess)
43	- Filters to code examples + API signatures + key notes
44	- Returns only essential content (205 tokens to Claude)
45
46	## Parameters
47
48	### Basic Usage
49
50	```bash
51	bash scripts/fetch-docs.sh [OPTIONS]
52	```
53
54	Required (pick one):
55	- `--library <name>` - Library name (e.g., "react", "nextjs")
56	- `--library-id <id>` - Direct Context7 ID (faster, skips resolution)
57
58	Optional:
59	- `--topic <topic>` - Specific feature to focus on
60	- `--mode <code\|info>` - code for examples (default), info for concepts
61	- `--page <1-10>` - Pagination for more results
62	- `--verbose` - Show token savings statistics
63
64	### Mode Selection
65
66	Code Mode (default): Returns code examples + API signatures
67	```bash
68	--mode code
69	```
70
71	Info Mode: Returns conceptual explanations + fewer examples
72	```bash
73	--mode info
74	```
75
76	## Common Library IDs
77
78	Use `--library-id` for faster lookup (skips resolution):
79
80	```bash
81	React: /reactjs/react.dev
82	Next.js: /vercel/next.js
83	Express: /expressjs/express
84	Prisma: /prisma/docs
85	MongoDB: /mongodb/docs
86	Fastify: /fastify/fastify
87	NestJS: /nestjs/docs
88	Vue.js: /vuejs/docs
89	Svelte: /sveltejs/site
90	```
91
92	## Workflow Patterns
93
94	### Pattern 1: Quick Code Examples
95
96	User asks: "Show me React useState examples"
97
98	```bash
99	bash scripts/fetch-docs.sh --library react --topic useState --verbose
100	```
101
102	Returns: 5 code examples + API signatures + notes (~205 tokens)
103
104	### Pattern 2: Learning New Library
105
106	User asks: "How do I get started with Prisma?"
107
108	```bash
109	# Step 1: Get overview
110	bash scripts/fetch-docs.sh --library prisma --topic "getting started" --mode info
111
112	# Step 2: Get code examples
113	bash scripts/fetch-docs.sh --library prisma --topic queries --mode code
114	```
115
116	### Pattern 3: Specific Feature Lookup
117
118	User asks: "How does Next.js routing work?"
119
120	```bash
121	bash scripts/fetch-docs.sh --library-id /vercel/next.js --topic routing
122	```
123
124	Using `--library-id` is faster when you know the exact ID.
125
126	### Pattern 4: Deep Exploration
127
128	User needs comprehensive information:
129
130	```bash
131	# Page 1: Basic examples
132	bash scripts/fetch-docs.sh --library react --topic hooks --page 1
133
134	# Page 2: Advanced patterns
135	bash scripts/fetch-docs.sh --library react --topic hooks --page 2
136	```
137
138	## Token Efficiency
139
140	How it works:
141
142	1. `fetch-docs.sh` calls `fetch-raw.sh` (which uses `mcp-client.py`)
143	2. Full response (934 tokens) stays in subprocess memory
144	3. Shell filters (awk/grep/sed) extract essentials (0 LLM tokens used)
145	4. Returns filtered output (205 tokens) to Claude
146
147	Savings:
148	- Direct MCP: 934 tokens per query
149	- This approach: 205 tokens per query
150	- 77% reduction
151
152	Do NOT use `mcp-client.py` directly - it bypasses filtering and wastes tokens.
153
154	## Advanced: Library Resolution
155
156	If library name fails, try variations:
157
158	```bash
159	# Try different formats
160	--library "next.js" # with dot
161	--library "nextjs" # without dot
162	--library "next" # short form
163
164	# Or search manually
165	bash scripts/fetch-docs.sh --library "your-library" --verbose
166	# Check output for suggested library IDs
167	```
168
169	## Verification
170
171	Run: `python3 scripts/verify.py`
172
173	Expected: `✓ fetch-docs.sh ready`
174
175	## If Verification Fails
176
177	1. Run diagnostic: `ls -la scripts/fetch-docs.sh`
178	2. Check: Script exists and is executable
179	3. Fix: `chmod +x scripts/fetch-docs.sh`
180	4. Stop and report if still failing - do not proceed wi

People who liked this also viewed

6 picks

Type