quickstart

Name: quickstart
Author: pinecone-io

23stars1forks

How to install quickstart?

pinecone-io/gemini-cli-extension/quickstart

$npx -y skills add pinecone-io/gemini-cli-extension --skill quickstart

Installs into the current project.

Summary

Interactive Pinecone quickstart for new developers. Choose between two paths - Database (create an integrated index, upsert data, and query using Pinecone MCP + Python) or Assistant (create a Pinecone Assistant for document Q&A). Use when a user wants to get started with Pinecone

Files · 1

View on GitHub

SKILL.md

1	# Pinecone Quickstart
2
3	Welcome! This skill walks you through your first Pinecone experience using the tools available to you. In this quickstart,
4	you will learn how to do a simple form of semantic search over some example data.
5
6	## Prerequisites
7
8	Before starting either path, verify the API key works by calling `list-indexes` via the Pinecone MCP. If it succeeds, proceed. If it fails, ask the user to set their key:
9
10	- Terminal: `export PINECONE_API_KEY="your-key"`
11	- Or create a `.env` file in the project root: `PINECONE_API_KEY=your-key`
12
13	Then retry `list-indexes` to confirm.
14
15	## Step 0: Choose Your Path
16
17	Ask the user which path they want:
18
19	- Database – Build a vector search index. Best for developers who want to store and search embeddings. Uses the Pinecone MCP + a Python upsert script.
20	- Assistant – Build a document Q&A assistant. Best for users who want to upload files and ask questions with cited answers. No code required.
21
22	---
23
24	## Path A: Database Quickstart
25
26	For each step, explain to the user what will happen. An overview is here:
27
28	1. Check if MCP is set
29	2. Create an integrated index with MCP
30	3. Upsert sample data using the bundled script (9 sentences across productivity, health, and nature themes)
31	4. Run a semantic search query and explore further queries
32	5. Optionally try reranking
33	6. Offer the complete standalone script
34
35	### Step 1 – Verify MCP is Available
36
37	The prerequisite check already called `list-indexes`. If it succeeded, the MCP is working — proceed to Step 2.
38
39	If it failed because MCP tools were unavailable (not an auth error):
40	- Tell the user the MCP server needs to be configured
41	- Point them to: https://docs.pinecone.io/reference/tools/mcp
42
43	### Step 2 – Create an Integrated Index
44
45	Use the MCP `create-index-for-model` tool to create a serverless index with integrated embeddings:
46
47	```
48	name: quickstart-skills
49	cloud: aws
50	region: us-east-1
51	embed:
52	model: llama-text-embed-v2
53	fieldMap:
54	text: chunk_text
55	```
56
57	Explain to the user what's happening:
58	- An integrated index uses a built-in Pinecone embedding model (`llama-text-embed-v2`)
59	- This means you send plain text and Pinecone handles the embedding automatically
60	- The `field_map` tells Pinecone which field in your records contains the text to embed
61
62	Wait for the index to become ready before proceeding. Waiting a few seconds is sufficient.
63
64	### Step 3 – Upsert Sample Data
65
66	Run the bundled upsert script to seed the index with sample records.
67
68	If `PINECONE_API_KEY` is set in the environment:
69	```bash
70	uv run scripts/upsert.py --index quickstart-skills
71	```
72
73	If using a `.env` file:
74	```bash
75	uv run --env-file .env scripts/upsert.py --index quickstart-skills
76	```
77
78	Explain to the user what's happening:
79	- The script uploads 9 sample records across three themes: productivity (getting work done), health (feeling unwell), and nature (outdoors/wildlife)
80	- The dataset is intentionally varied so semantic search can show its value — the queries below use completely different words than the records, but the right ones still surface
81	- Each record has an `_id`, a `chunk_text` field (the text that gets embedded), and a `category` field
82	- This is the same structure you'd use for your own data — just replace the records
83
84	### Step 4 – Query with the MCP
85
86	Use the MCP `search-records` tool to run the first semantic search:
87
88	```
89	index: quickstart-skills
90	namespace: example-namespace
91	query:
92	topK: 3
93	inputs:
94	text: "getting things done efficiently"
95	```
96
97	Display the results in a clean table: ID, score, and `chunk_text`.
98
99	Explain to the user what's happening:
100	- Notice the query shares no keywords with the records — but it surfaces the productivity sentences
101	- That's semantic search: it finds meaning, not just matching words
102	- You sent plain text — Pinecone embedded the query using the same model as the index
103
104	Offer to explore further: Ask the user if they'd like to try another query to see the effect more clearly:
105	- Option A: `"feeling under the weather"` — should surface the health records
106	- Option B: `"wildlife spotting outside"` — should surface the nature records
107	- Option C: No thanks, move on
108
109	Run whichever query they choose and display the results the same way. If they want to try both, do both. After each result, point out which theme surfaced and why.
110
111	If they decline or are done exploring, proceed to Step 5 or offer to skip ahead to the complete script.
112
113	### Step 5 – Try Reranking (Optional)
114
115	Ask the user if they want to try reranking.
116
117	If yes, use `search-records` again with reranking enabled:
118
119	```
120	rerank:
121	model: bge-reranker-v2-m3
122	rankFields

People who liked this also viewed

6 picks

Type

quickstart

23stars1forks

View on GitHub

How to install quickstart?

pinecone-io/gemini-cli-extension/quickstart

$npx -y skills add pinecone-io/gemini-cli-extension --skill quickstart

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Pinecone Quickstart
2
3	Welcome! This skill walks you through your first Pinecone experience using the tools available to you. In this quickstart,
4	you will learn how to do a simple form of semantic search over some example data.
5
6	## Prerequisites
7
8	Before starting either path, verify the API key works by calling `list-indexes` via the Pinecone MCP. If it succeeds, proceed. If it fails, ask the user to set their key:
9
10	- Terminal: `export PINECONE_API_KEY="your-key"`
11	- Or create a `.env` file in the project root: `PINECONE_API_KEY=your-key`
12
13	Then retry `list-indexes` to confirm.
14
15	## Step 0: Choose Your Path
16
17	Ask the user which path they want:
18
19	- Database – Build a vector search index. Best for developers who want to store and search embeddings. Uses the Pinecone MCP + a Python upsert script.
20	- Assistant – Build a document Q&A assistant. Best for users who want to upload files and ask questions with cited answers. No code required.
21
22	---
23
24	## Path A: Database Quickstart
25
26	For each step, explain to the user what will happen. An overview is here:
27
28	1. Check if MCP is set
29	2. Create an integrated index with MCP
30	3. Upsert sample data using the bundled script (9 sentences across productivity, health, and nature themes)
31	4. Run a semantic search query and explore further queries
32	5. Optionally try reranking
33	6. Offer the complete standalone script
34
35	### Step 1 – Verify MCP is Available
36
37	The prerequisite check already called `list-indexes`. If it succeeded, the MCP is working — proceed to Step 2.
38
39	If it failed because MCP tools were unavailable (not an auth error):
40	- Tell the user the MCP server needs to be configured
41	- Point them to: https://docs.pinecone.io/reference/tools/mcp
42
43	### Step 2 – Create an Integrated Index
44
45	Use the MCP `create-index-for-model` tool to create a serverless index with integrated embeddings:
46
47	```
48	name: quickstart-skills
49	cloud: aws
50	region: us-east-1
51	embed:
52	model: llama-text-embed-v2
53	fieldMap:
54	text: chunk_text
55	```
56
57	Explain to the user what's happening:
58	- An integrated index uses a built-in Pinecone embedding model (`llama-text-embed-v2`)
59	- This means you send plain text and Pinecone handles the embedding automatically
60	- The `field_map` tells Pinecone which field in your records contains the text to embed
61
62	Wait for the index to become ready before proceeding. Waiting a few seconds is sufficient.
63
64	### Step 3 – Upsert Sample Data
65
66	Run the bundled upsert script to seed the index with sample records.
67
68	If `PINECONE_API_KEY` is set in the environment:
69	```bash
70	uv run scripts/upsert.py --index quickstart-skills
71	```
72
73	If using a `.env` file:
74	```bash
75	uv run --env-file .env scripts/upsert.py --index quickstart-skills
76	```
77
78	Explain to the user what's happening:
79	- The script uploads 9 sample records across three themes: productivity (getting work done), health (feeling unwell), and nature (outdoors/wildlife)
80	- The dataset is intentionally varied so semantic search can show its value — the queries below use completely different words than the records, but the right ones still surface
81	- Each record has an `_id`, a `chunk_text` field (the text that gets embedded), and a `category` field
82	- This is the same structure you'd use for your own data — just replace the records
83
84	### Step 4 – Query with the MCP
85
86	Use the MCP `search-records` tool to run the first semantic search:
87
88	```
89	index: quickstart-skills
90	namespace: example-namespace
91	query:
92	topK: 3
93	inputs:
94	text: "getting things done efficiently"
95	```
96
97	Display the results in a clean table: ID, score, and `chunk_text`.
98
99	Explain to the user what's happening:
100	- Notice the query shares no keywords with the records — but it surfaces the productivity sentences
101	- That's semantic search: it finds meaning, not just matching words
102	- You sent plain text — Pinecone embedded the query using the same model as the index
103
104	Offer to explore further: Ask the user if they'd like to try another query to see the effect more clearly:
105	- Option A: `"feeling under the weather"` — should surface the health records
106	- Option B: `"wildlife spotting outside"` — should surface the nature records
107	- Option C: No thanks, move on
108
109	Run whichever query they choose and display the results the same way. If they want to try both, do both. After each result, point out which theme surfaced and why.
110
111	If they decline or are done exploring, proceed to Step 5 or offer to skip ahead to the complete script.
112
113	### Step 5 – Try Reranking (Optional)
114
115	Ask the user if they want to try reranking.
116
117	If yes, use `search-records` again with reranking enabled:
118
119	```
120	rerank:
121	model: bge-reranker-v2-m3
122	rankFields

People who liked this also viewed

6 picks

Type