### Pinecone FTS Index Setup and Search

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-full-text-search/SKILL.md

This Python script demonstrates the end-to-end process of setting up a Pinecone index for full-text search with filterable metadata. It includes schema definition, index creation, document upsertion, and a search query combining text relevance and metadata filtering. Ensure you have the Pinecone client installed and your API key configured.

```python
import time
from pinecone import Pinecone
from pinecone.preview import SchemaBuilder

INDEX_NAME = "my-fts-index"        # TODO: name your index (lowercase alphanumeric + hyphens, ≤45 chars)
NAMESPACE = "__default__"          # TODO: pick a namespace; auto-created on first upsert

pc = Pinecone()                    # reads PINECONE_API_KEY
# TODO: preprod backends require an x-environment header on the client:
#   pc = Pinecone(additional_headers={"x-environment": "preprod-aws-0"})

# 1. Schema — one FTS string field, one filterable string, one filterable float.
#    Field names must NOT start with `_` (reserved for `_id` / `_score`) or `$`
#    (reserved for filter operators), and are limited to 64 bytes.
schema = (
    SchemaBuilder()
    .add_string_field("body", full_text_search={"language": "en"})  # TODO: rename for your content
    .add_string_field("category", filterable=True)                   # TODO: any exact-match metadata
    .add_integer_field("year", filterable=True)                      # TODO: any numeric filter — emits `"type": "float"` on the wire
    .build()
)

# 2. Create the index. read_capacity defaults to {"mode": "OnDemand"}; pass
#    {"mode": "Dedicated", ...} only if you specifically want provisioned reads.
if not pc.preview.indexes.exists(INDEX_NAME):
    pc.preview.indexes.create(name=INDEX_NAME, schema=schema)

# 3. Wait for the index itself to become Ready.
while not pc.preview.indexes.describe(INDEX_NAME).status.ready:
    time.sleep(5)

idx = pc.preview.index(name=INDEX_NAME)

# 4. Upsert a single document. `_id` is required, every other field is optional.
#    upsert REPLACES the document on conflict — there is no per-field merge in 2026-01.alpha.
idx.documents.upsert(
    namespace=NAMESPACE,
    documents=[
        {
            "_id": "doc-1",
            "body": "Full-text search is great for keyword queries.",
            "category": "intro",
            "year": 2025.0,
        }
    ],
)

# 5. Poll until the FTS side is searchable (upsert returns BEFORE docs are indexed).
deadline = time.time() + 300
while time.time() < deadline:
    resp = idx.documents.search(
        namespace=NAMESPACE, top_k=1,
        score_by=[{"type": "text", "field": "body", "query": "search"}],  # TODO: sentinel query likely to hit
        include_fields=[],          # required on every search; [] = lightest payload (ids + _score only)
    )
    if resp.matches:
        break
    time.sleep(5)

# 6. Search — text scoring composed with metadata filter.
resp = idx.documents.search(
    namespace=NAMESPACE,
    top_k=5,
    score_by=[{"type": "text", "field": "body", "query": "keyword queries"}],
    filter={"year": {"$gte": 2024}},        # TODO: adjust filter or drop it
    include_fields=["*"],                    # "*" = all stored fields; [] = `_id` + `_score` only
)
for m in resp.matches:
    print(m._id, getattr(m, "_score", getattr(m, "score", None)), m.to_dict())

```

--------------------------------

### Install Pinecone Skill Library

Source: https://context7.com/pinecone-io/skills/llms.txt

Install the Pinecone Agent Skills library using npm.

```bash
# Install the skill library
npx skills add pinecone-io/skills
```

--------------------------------

### Copy Quickstart Script

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-quickstart/SKILL.md

Copies a complete Python quickstart script to the current working directory. This script handles index creation, upserting, querying, and reranking.

```bash
cp scripts/quickstart_complete.py ./pinecone_quickstart.py
```

--------------------------------

### Multi-Step Request Example

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-assistant/SKILL.md

Demonstrates a multi-step request involving creating an assistant, uploading files, and then asking a question.

```bash
uv run scripts/create.py --name docs-bot
uv run scripts/upload.py --assistant docs-bot --source ./docs
uv run scripts/chat.py --assistant docs-bot --message "what are the main features?"
```

--------------------------------

### Install uv

Source: https://context7.com/pinecone-io/skills/llms.txt

Install uv, a fast Python package installer and virtual environment manager, used for running bundled Python scripts.

```bash
# Optional: uv (runs bundled Python scripts)
# https://docs.astral.sh/uv/getting-started/installation/
```

--------------------------------

### Example Upload Command

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-assistant/references/upload.md

Use this command to upload files from a directory to a specified assistant. You can customize patterns to include or exclude specific file types.

```bash
uv run scripts/upload.py \
  --assistant "assistant-name" \
  --source "./docs" \
  --patterns "*.md,*.pdf"
```

--------------------------------

### Install Pinecone CLI

Source: https://context7.com/pinecone-io/skills/llms.txt

Install the Pinecone CLI tool using Homebrew. This enables the `pinecone-cli` skill.

```bash
# Optional: Pinecone CLI (enables pinecone-cli skill)
brew tap pinecone-io/tap && brew install pinecone-io/tap/pinecone
```

--------------------------------

### Metadata Example

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-assistant/references/upload.md

Include additional metadata in JSON format to enrich your uploaded documents. This can be useful for tracking sources or versions.

```bash
--metadata '{"source":"github","repo":"owner/repo","branch":"main"}'
```

--------------------------------

### Perform Semantic Search with MCP

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-quickstart/SKILL.md

Query your Pinecone index using the MCP `search-records` tool. This example demonstrates semantic search by querying for "getting things done efficiently", which retrieves relevant text embeddings.

```yaml
index: quickstart-skills
namespace: example-namespace
query:
  topK: 3
  inputs:
    text: "getting things done efficiently"
```

--------------------------------

### Install Pinecone CLI on macOS

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-cli/SKILL.md

Use Homebrew to install the Pinecone CLI on macOS. Download from GitHub Releases for Linux and Windows.

```bash
brew tap pinecone-io/tap
brew install pinecone-io/tap/pinecone
```

--------------------------------

### Upload a directory with patterns

Source: https://context7.com/pinecone-io/skills/llms.txt

Uploads files from a local directory to a Pinecone assistant, filtering by specified patterns. Ensure the 'uv' command-line tool is installed and configured.

```bash
uv run scripts/upload.py \
  --assistant docs-bot \
  --source ./docs \
  --patterns "*.md,*.pdf"
```

--------------------------------

### Install Pinecone CLI

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-help/SKILL.md

Install the Pinecone CLI tool using Homebrew. This command allows you to manage Pinecone indexes and perform batch operations from your terminal.

```bash
brew tap pinecone-io/tap && brew install pinecone-io/tap/pinecone
```

--------------------------------

### Install Pinecone CLI on macOS

Source: https://context7.com/pinecone-io/skills/llms.txt

Install the Pinecone CLI using Homebrew on macOS. Ensure you have Homebrew installed before running these commands.

```bash
# macOS
brew tap pinecone-io/tap
brew install pinecone-io/tap/pinecone
```

--------------------------------

### describe-index

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-mcp/SKILL.md

Get configuration details for a specific index.

```APIDOC
## describe-index

### Description
Get configuration details for a specific index — cloud, region, dimension, metric, embedding model, field map, and status.

### Method
GET

### Endpoint
/describe-index

### Parameters
#### Query Parameters
- **name** (string) - Required - Index name
```

--------------------------------

### List, Describe, and Get Stats for Pinecone Indexes

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-cli/references/command-reference.md

Commands to view a summary or detailed information about your Pinecone indexes, including host, embed, and tags. JSON output is available for programmatic use.

```bash
pc index list                          # Summary view
```

```bash
pc index list --wide                   # Additional columns (host, embed, tags)
```

```bash
pc index list -j                       # JSON output
```

```bash
pc index describe -n my-index
```

```bash
pc index describe -n my-index -j
```

```bash
pc index stats -n my-index
```

```bash
pc index stats -n my-index --filter '{"genre":{"$eq":"rock"}}'
```

--------------------------------

### Pinecone Search Query Example

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-full-text-search/references/onboarding-walkthrough.md

Example of a Pinecone documents search query using score_by for text relevance and including all fields. Refer to querying documentation for variations.

```python
idx.documents.search(
    score_by=[{"type":"text", "field":"<fts_field>", "query":"<sentinel>"}],
    include_fields=["*"],
    top_k=3
)
```

--------------------------------

### Example Record for Upsert

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-mcp/SKILL.md

This is an example of a record structure expected by the `upsert-records` tool. Ensure records have an `_id` or `id` and include the text field specified in the index's `fieldMap`.

```json
{
  "_id": "rec1",
  "chunk_text": "The Eiffel Tower was built in 1889.",
  "category": "architecture"
}
```

--------------------------------

### Execute Chat with Assistant Command

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-assistant/references/chat.md

Use this command to send a message to a specific assistant and get a response. Ensure the assistant name and message are provided.

```bash
uv run scripts/chat.py \
  --assistant "assistant-name" \
  --message "user's question"
```

--------------------------------

### Add String Field with Full-Text Search Configuration

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-full-text-search/references/schema-design.md

Configure a string field to support full-text search. This example shows how to enable English language search with stemming enabled, and includes a description for LLM-driven workflows.

```python
.add_string_field(
    "body",
    full_text_search={"language": "en", "stemming": True},
    description="Full article text. Use for keyword searches, narrative phrases, and topical queries.",
)
```

--------------------------------

### Multi-step chained request: create, upload, chat

Source: https://context7.com/pinecone-io/skills/llms.txt

Demonstrates a multi-step process involving creating an assistant, uploading data to it, and then querying it with a chat message. This showcases sequential command execution.

```bash
uv run scripts/create.py --name docs-bot
uv run scripts/upload.py --assistant docs-bot --source ./docs
uv run scripts/chat.py --assistant docs-bot --message "What are the main features?"
```

--------------------------------

### Pinecone CLI Best Practices for Authentication

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-cli/references/troubleshooting.md

Use interactive login for development, service accounts for CI/CD, and API keys for quick testing.

```bash
pc login
pc api-key create -n "my-key" --store
```

--------------------------------

### Create a Pinecone Assistant

Source: https://context7.com/pinecone-io/skills/llms.txt

Create a Pinecone Assistant for document Q&A. This command sets up the assistant with a name, instructions, and region. It supports using a `.env` file for configuration.

```bash
uv run scripts/create.py \
  --name docs-bot \
  --instructions "Use professional technical tone and cite sources." \
  --region us

# With .env file
uv run --env-file .env scripts/create.py --name docs-bot
```

--------------------------------

### List API Keys

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-cli/references/command-reference.md

Lists all available API keys.

```bash
pc api-key list
```

--------------------------------

### List Projects

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-cli/references/command-reference.md

Lists all available Pinecone projects.

```bash
pc project list
```

--------------------------------

### Apply Sync

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-assistant/references/sync.md

Execute the sync script to upload new or changed files to the assistant.

```bash
uv run scripts/sync.py --assistant my-docs --source ./docs
```

--------------------------------

### describe-index-stats

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-mcp/SKILL.md

Get statistics for an index including total record count and per-namespace breakdown.

```APIDOC
## describe-index-stats

### Description
Get statistics for an index including total record count and per-namespace breakdown.

### Method
GET

### Endpoint
/describe-index-stats

### Parameters
#### Query Parameters
- **name** (string) - Required - Index name
```

--------------------------------

### Get Pinecone Index Statistics

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-cli/SKILL.md

Command to retrieve statistics for a specific Pinecone index.

```bash
pc index stats -n my-index
```

--------------------------------

### Create Project

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-cli/references/command-reference.md

Creates a new Pinecone project with a specified name. The `--target` flag can be used to immediately switch to the newly created project.

```bash
pc project create -n "demo-project"
```

```bash
pc project create -n "demo-project" --target
```

--------------------------------

### Execute Sync Script

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-assistant/references/sync.md

Run the sync script with specified assistant and source directories. Optional flags include `--delete-missing`, `--dry-run`, and `--yes` for automation.

```bash
uv run scripts/sync.py \
  --assistant "assistant-name" \
  --source "./docs" \
  [--delete-missing] \
  [--dry-run] \
  [--yes]
```

--------------------------------

### Add Pinecone Skills

Source: https://github.com/pinecone-io/skills/blob/main/README.md

Use this command to add the Pinecone Agent Skills to your project. Ensure you have Node.js and npm installed.

```bash
npx skills add pinecone-io/skills
```

--------------------------------

### Preview Sync Changes

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-assistant/references/sync.md

Use the `--dry-run` flag to preview changes without applying them. This is recommended before executing the actual sync.

```bash
uv run scripts/sync.py --assistant my-docs --source ./docs --dry-run
```

--------------------------------

### Create and Export API Key for SDKs

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-cli/SKILL.md

Generate an API key using the CLI and export it as an environment variable for use with Pinecone SDKs. This is necessary because `pc login` does not set `PINECONE_API_KEY`.

```bash
KEY=$(pc api-key create --name agent-sdk-key --json | jq -r '.value')
export PINECONE_API_KEY="$KEY"
```

--------------------------------

### List Pinecone Assistants

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-assistant/references/list.md

Use this command to list all assistants in your account. The `--files` flag shows file details, and `--json` provides output in JSON format.

```bash
# Basic listing
uv run scripts/list.py
```

```bash
# With file details
uv run scripts/list.py --files
```

```bash
# JSON output
uv run scripts/list.py --json
```

```bash
# JSON with files (useful for scripting)
uv run scripts/list.py --files --json
```

--------------------------------

### Create API Key

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-cli/references/command-reference.md

Creates a new API key with a specified name. Options include storing the key locally or associating it with a specific project.

```bash
pc api-key create -n "my-key"
```

```bash
pc api-key create -n "my-key" --store
```

```bash
pc api-key create -n "my-key" -i proj-abc123
```

--------------------------------

### Get Context Snippets from Pinecone Assistant

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-assistant/SKILL.md

Retrieve context snippets from a Pinecone Assistant based on a query. Specify the number of top results with --top-k.

```bash
uv run scripts/context.py --assistant <assistant-name> --query "<your-query>" --top-k <number-of-results>
```

--------------------------------

### Create, List, Describe, and Restore Pinecone Backups

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-cli/references/command-reference.md

Manage backups for your Pinecone indexes. Create new backups, list existing ones, describe backup details, and restore indexes from backups. Supports enabling deletion protection on restored indexes.

```bash
# Create / list / describe
pc backup create -i my-index -n "nightly-backup" -d "Backup before deployment"
```

```bash
pc backup list
```

```bash
pc backup list --index-name my-index
```

```bash
pc backup describe -i <backup-uuid>
```

```bash
# Restore (creates a new index)
pc backup restore -i <backup-uuid> -n restored-index
```

```bash
pc backup restore -i <backup-uuid> -n restored-index --deletion-protection enabled
```

```bash
# Check restore job status
pc backup restore list
```

```bash
pc backup restore describe -i rj-abc123
```

```bash
# Delete backup
pc backup delete -i <backup-uuid>
```

--------------------------------

### Get Pinecone Index Statistics with Filter

Source: https://context7.com/pinecone-io/skills/llms.txt

Retrieve statistics for a Pinecone index, applying a filter to include only data matching specific criteria. The filter is applied as a JSON string.

```bash
# Stats with filter
pc index stats -n my-index --filter '{"genre":{"$eq":"rock"}}'
```

--------------------------------

### Create Pinecone Index Backup

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-cli/SKILL.md

Create a backup (snapshot) of a Pinecone index with a specified name.

```bash
pc backup create -i my-index -n "my-backup"
```

--------------------------------

### Describe Project

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-cli/references/command-reference.md

Displays detailed information about a specific Pinecone project using its ID.

```bash
pc project describe -i proj-abc123
```

--------------------------------

### Automate CI/CD with Pinecone CLI

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-cli/SKILL.md

Configure authentication using client ID and secret for CI/CD environments and perform bulk vector upserts with a specified batch size.

```bash
export PINECONE_CLIENT_ID="..."
export PINECONE_CLIENT_SECRET="..."
pc auth configure --client-id "$PINECONE_CLIENT_ID" --client-secret "$PINECONE_CLIENT_SECRET"
pc index vector upsert -n my-index --file ./vectors.jsonl --batch-size 1000
```

--------------------------------

### Automate Pinecone Backups

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-cli/references/troubleshooting.md

Automate the creation of backups for your Pinecone indexes, including dynamic naming based on the current date.

```bash
pc backup create -i my-index -n "daily-backup-$(date +%Y%m%d)"
```

--------------------------------

### List assistants

Source: https://context7.com/pinecone-io/skills/llms.txt

Lists all available assistants with their basic information in a table format. This command provides a quick overview of the assistants managed by the system.

```bash
uv run scripts/list.py
```

--------------------------------

### Apply Sync and Delete Missing

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-assistant/references/sync.md

Apply the sync and remove files from the assistant that are no longer present locally. This is useful after cleaning up local files or after a `git pull`.

```bash
uv run scripts/sync.py --assistant my-docs --source ./docs --delete-missing
```

--------------------------------

### Create Pinecone Assistant

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-quickstart/SKILL.md

Invokes the Pinecone Assistant script to create a new assistant. Use `--env-file .env` if your API key is in a .env file.

```bash
uv run ../pinecone-assistant/scripts/create.py --name my-assistant
```

--------------------------------

### List assistants with file counts

Source: https://context7.com/pinecone-io/skills/llms.txt

Lists all assistants and includes the file count associated with each. This provides more detailed information for managing assistant data.

```bash
uv run scripts/list.py --files
```

--------------------------------

### Execute Context Retrieval

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-assistant/references/context.md

Run the context retrieval script with specified assistant, query, and top-k parameters. Ensure the SDK is updated to v8.0.0+ for this functionality.

```bash
uv run scripts/context.py \
  --assistant "assistant-name" \
  --query "search text" \
  --top-k 5
```

--------------------------------

### Run Skill Linting Checks

Source: https://github.com/pinecone-io/skills/blob/main/CLAUDE.md

Execute these commands from the ../tools/ directory to lint skills, check source tags, and validate links within the skills directory.

```bash
uv run ../tools/check-skills.py --skills-dir skills/
```

```bash
uv run ../tools/check-source-tags.py --dir .
```

```bash
uv run ../tools/check-links.py --skills-dir skills/
```

--------------------------------

### Create Pinecone Assistant

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-assistant/SKILL.md

Use this script to create a new Pinecone Assistant. Specify the assistant's name, instructions, and region.

```bash
uv run scripts/create.py --name <assistant-name> --instructions "<your-instructions>" --region <aws-region>
```

--------------------------------

### Create a Serverless Pinecone Index

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-cli/SKILL.md

Command to create a new serverless Pinecone index with specified name, dimension, metric, cloud provider, and region.

```bash
pc index create -n my-index -d 1536 -m cosine -c aws -r us-east-1
```

--------------------------------

### Backup and Restore Pinecone Index

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-cli/SKILL.md

Create a snapshot backup of an index before a migration and restore it to a new index if necessary. Requires backup UUID for restore.

```bash
# Snapshot before a migration
pc backup create -i my-index -n "pre-migration"
```

```bash
# Restore to a new index if something goes wrong
pc backup restore -i <backup-uuid> -n my-index-restored
```

--------------------------------

### Chat with Pinecone Assistant

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-quickstart/SKILL.md

Initiates a chat with a Pinecone Assistant, sending a message and receiving answers with citations. Use `--env-file .env` if needed.

```bash
uv run ../pinecone-assistant/scripts/chat.py --assistant my-assistant --message "What are the main topics in these documents?"
```

--------------------------------

### Bulk ingest data using a script

Source: https://context7.com/pinecone-io/skills/llms.txt

Uses a bundled script for recommended bulk data loading into a Pinecone index. It handles batching, error inspection, and readiness polling. Ensure the data file is in JSONL format.

```bash
# Recommended for bulk loads — handles batch_upsert + error inspection + readiness polling
uv run --script scripts/ingest.py \
  --data processed.jsonl \
  --index my-fts-index \
  --sentinel-field body \
  --batch-size 50 \
  --namespace __default__
```

--------------------------------

### Upsert Sample Records into Integrated Index

Source: https://context7.com/pinecone-io/skills/llms.txt

Use this script to upsert sample records into a pre-created integrated index. Ensure the index exists and the PINECONE_API_KEY environment variable is set.

```bash
# Step 1 – Upsert sample records into a pre-created integrated index
uv run scripts/upsert.py --index quickstart-skills

# Expected output:
# Upserted 9 records into 'quickstart-skills' (namespace: 'example-namespace')
```

--------------------------------

### Create Assistant CLI Command

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-assistant/references/create.md

Use this command to create a new Pinecone Assistant. Specify a unique name, optional instructions for behavior, and the desired region (us or eu).

```bash
uv run scripts/create.py \
  --name "assistant-name" \
  --instructions "instructions" \
  --region "us"
```

--------------------------------

### List Pinecone Indexes

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-cli/SKILL.md

Command to list all available Pinecone indexes.

```bash
pc index list
```

--------------------------------

### List Namespaces in Index

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-cli/SKILL.md

View all namespaces present within a specific Pinecone index.

```bash
pc index namespace list -n my-index
```

--------------------------------

### Create Serverless Index with Integrated Embeddings

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-quickstart/SKILL.md

Use the Pinecone MCP `create-index-for-model` tool to set up a serverless index. This configuration uses an integrated embedding model, automatically handling text embedding.

```yaml
name: quickstart-skills
cloud: aws
region: us-east-1
embed:
  model: llama-text-embed-v2
  fieldMap:
    text: chunk_text
```

--------------------------------

### Manage Pinecone Target Context

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-cli/references/troubleshooting.md

Troubleshoot 'Project not found' or 'Organization not found' errors by showing, clearing, or setting the target context.

```bash
pc target --show
pc target --clear
pc target -o "my-org" -p "my-project"
```

--------------------------------

### Manage Pinecone API Key

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-cli/references/troubleshooting.md

Check the currently configured API key and create a new one if the existing key is not working. New keys should be stored for use.

```bash
pc config get-api-key
pc api-key create -n "new-key" --store
```

--------------------------------

### Backup and Restore

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-cli/references/command-reference.md

Commands for managing backups of Pinecone indexes, including creating, listing, describing, restoring, and deleting backups.

```APIDOC
## Backup Operations

### Description
Manages backups for Pinecone indexes, enabling data protection and disaster recovery.

### Create Backup
#### Usage
```bash
pc backup create -i <index-name> -n <backup-name> -d "<description>"
```
#### Parameters
- `-i`, `--index-name` (string): The name of the index to back up.
- `-n`, `--name` (string): The name for the backup.
- `-d`, `--description` (string): A description for the backup.

### List Backups
#### Usage
```bash
pc backup list [--index-name <index-name>]
```
#### Parameters
- `--index-name` (string): Optional. Filter backups by index name.

### Describe Backup
#### Usage
```bash
pc backup describe -i <backup-uuid>
```
#### Parameters
- `-i`, `--backup-uuid` (string): The UUID of the backup to describe.

### Restore Backup
#### Usage
```bash
pc backup restore -i <backup-uuid> -n <new-index-name> [--deletion-protection <enabled|disabled>]
```
#### Parameters
- `-i`, `--backup-uuid` (string): The UUID of the backup to restore.
- `-n`, `--name` (string): The name for the new index created from the backup.
- `--deletion-protection` (string): Enable or disable deletion protection for the restored index.

### List / Describe Restore Jobs
#### Usage
```bash
pc backup restore list
pc backup restore describe -i <restore-job-uuid>
```
#### Parameters
- `-i`, `--restore-job-uuid` (string): The UUID of the restore job to describe.

### Delete Backup
#### Usage
```bash
pc backup delete -i <backup-uuid>
```
#### Parameters
- `-i`, `--backup-uuid` (string): The UUID of the backup to delete.
```

--------------------------------

### Preview file sync operations

Source: https://context7.com/pinecone-io/skills/llms.txt

Performs a dry run of the file synchronization process, showing what changes would be made without applying them. This helps in verifying the sync operation before execution.

```bash
uv run scripts/sync.py --assistant docs-bot --source ./docs --dry-run
```

--------------------------------

### Sync Files with Pinecone Assistant

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-assistant/SKILL.md

Perform an incremental sync of files with a Pinecone Assistant. Use --delete-missing to remove files not present in the source and --dry-run to preview changes.

```bash
uv run scripts/sync.py --assistant <assistant-name> --source <source-directory> --delete-missing --dry-run
```

--------------------------------

### Manage Namespaces in Pinecone Index

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-cli/references/command-reference.md

Commands for creating, listing, describing, and deleting namespaces within a Pinecone index. Namespaces allow for logical separation of vectors.

```bash
pc index namespace create -n my-index --name tenant-a
```

```bash
pc index namespace create -n my-index --name tenant-b --schema "category,brand"
```

```bash
pc index namespace list -n my-index
```

```bash
pc index namespace list -n my-index --prefix "tenant-"
```

```bash
pc index namespace describe -n my-index --name tenant-a
```

```bash
pc index namespace delete -n my-index --name tenant-a   # WARNING: deletes all vectors
```

--------------------------------

### MCP Tool: Describe Index Configuration

Source: https://context7.com/pinecone-io/skills/llms.txt

Use this MCP tool to retrieve details about a specific index, including its cloud provider, region, dimension, metric, embedding model, and status.

```text
# MCP tool: describe-index
name: quickstart-skills
# Returns cloud, region, dimension, metric, embedding model, fieldMap, status
```

--------------------------------

### Create Integrated Index, Upsert Records, and Perform Semantic Search

Source: https://context7.com/pinecone-io/skills/llms.txt

This Python script demonstrates creating an integrated index, upserting records with text fields, and performing semantic search. It requires the pinecone-python client and PINECONE_API_KEY.

```python
#!/usr/bin/env python3
# /// script
# dependencies = ["pinecone>=8.0.0"]
# ///
import os
from pinecone import Pinecone

pc = Pinecone(api_key=os.environ["PINECONE_API_KEY"])

# Create an integrated index (Pinecone handles embedding automatically)
if not pc.has_index("quickstart"):
    pc.create_index_for_model(
        name="quickstart",
        cloud="aws",
        region="us-east-1",
        embed={"model": "llama-text-embed-v2", "field_map": {"text": "chunk_text"}},
    )

records = [
    {"_id": "rec1", "chunk_text": "I've been sneezing all day and my nose won't stop running.", "category": "health"},
    {"_id": "rec4", "chunk_text": "She blocked off two hours in the morning to focus without interruptions.", "category": "productivity"},
    {"_id": "rec7", "chunk_text": "A red fox darted across the trail and disappeared into the underbrush.", "category": "nature"},
]

idx = pc.Index("quickstart")
idx.upsert_records("example-namespace", records)

# Semantic search — query words differ completely from record words
results = idx.search(
    namespace="example-namespace",
    query={"top_k": 3, "inputs": {"text": "feeling ill and run down"}},
)
for hit in results["result"]["hits"]:
    print(f"id={hit['_id']} score={round(hit['_score'], 2)} text={hit['fields']['chunk_text']}")

# Search with reranking
reranked = idx.search(
    namespace="example-namespace",
    query={"top_k": 3, "inputs": {"text": "feeling ill and run down"}},
    rerank={"model": "bge-reranker-v2-m3", "top_n": 3, "rank_fields": ["chunk_text"]},
)
```

--------------------------------

### Chat with an assistant

Source: https://context7.com/pinecone-io/skills/llms.txt

Initiates a chat session with a specified assistant, sending a message and receiving a response. This is useful for querying information contained within the assistant's data.

```bash
uv run scripts/chat.py \
  --assistant docs-bot \
  --message "What are the main features described in these documents?"
```

--------------------------------

### Verify Pinecone Target Context

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-cli/references/troubleshooting.md

Ensure the CLI is targeting the correct project, especially when using service accounts that might have restricted access.

```bash
pc target --show
```

--------------------------------

### Use JSON Output with Pinecone CLI for Scripting

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-cli/references/troubleshooting.md

Utilize JSON output with tools like `jq` for easier parsing and scripting of Pinecone CLI commands.

```bash
pc index list -j | jq -r '.[] | .name'
```

--------------------------------

### Upload Files to Pinecone Assistant

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-assistant/SKILL.md

Upload files to an existing Pinecone Assistant. Specify the assistant name, source directory, and file patterns.

```bash
uv run scripts/upload.py --assistant <assistant-name> --source <source-directory> --patterns "<file-patterns>"
```

--------------------------------

### Upload a directory with metadata tagging

Source: https://context7.com/pinecone-io/skills/llms.txt

Uploads files to a Pinecone assistant and tags them with custom metadata. This is useful for organizing and filtering uploaded data.

```bash
uv run scripts/upload.py \
  --assistant docs-bot \
  --source ./docs \
  --metadata '{"source":"github","repo":"owner/repo","branch":"main"}'
```

--------------------------------

### Bulk Metadata Update with Preview

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-cli/SKILL.md

Update metadata for multiple vectors based on a filter. Includes a `--dry-run` option to preview changes before applying them.

```bash
# Preview first
pc index vector update -n my-index \
  --filter '{"env":{"$eq":"staging"}}' \
  --metadata '{"env":"production"}' \
  --dry-run
```

```bash
# Apply
pc index vector update -n my-index \
  --filter '{"env":{"$eq":"staging"}}' \
  --metadata '{"env":"production"}'
```

--------------------------------

### Describe API Key

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-cli/references/command-reference.md

Displays detailed information about a specific API key using its ID.

```bash
pc api-key describe -i key-abc123
```

--------------------------------

### List Organizations

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-cli/references/command-reference.md

Lists all available Pinecone organizations.

```bash
pc organization list
```

--------------------------------

### Upload Documents to Assistant

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-quickstart/SKILL.md

Uploads documents from a specified source directory to a Pinecone Assistant. Pinecone automatically handles chunking, embedding, and indexing. Use `--env-file .env` if needed.

```bash
uv run ../pinecone-assistant/scripts/upload.py --assistant my-assistant --source ./your-docs
```

--------------------------------

### Create Pinecone Index

Source: https://github.com/pinecone-io/skills/blob/main/skills/pinecone-cli/references/command-reference.md

Use this command to create a new Pinecone index. Supports serverless and sparse vector indexes, with options for integrated embedding models, deletion protection, and creating from collections.

```bash
# Serverless index
pc index create -n my-index -d 1536 -m cosine -c aws -r us-east-1
```

```bash
# With integrated embedding model
pc index create -n my-index -m cosine -c aws -r us-east-1 \
  --model multilingual-e5-large \
  --field-map text=chunk_text
```

```bash
# Sparse vector index
pc index create -n sparse-index -m dotproduct -c aws -r us-east-1 --vector-type sparse
```

```bash
# With deletion protection
pc index create -n my-index -d 1536 -m cosine -c aws -r us-east-1 --deletion-protection enabled
```

```bash
# From collection
pc index create -n my-index -d 1536 -m cosine -c aws -r us-east-1 --source-collection my-collection
```