Pinecone
Pineconedatabase6 credentials
Pinecone API Key
pinecone / api-key
Pinecone API keys authenticate requests to Pinecone Data Plane and Control Plane APIs. The official quickstart and SDK examples use PINECONE_API_KEY, and the API reference documents API-key authentication for Pinecone requests.
Location
Api-Key, X-Pinecone-API-Key, AuthorizationPinecone REST, SDK, gateway, and proxy request authentication contexts
PINECONE_API_KEY.env, .env.local, .env.production, pinecone.yml, pinecone.yaml, config.yaml, config.yml, docker-compose.yml, compose.yml, values.yaml, deployment.yaml, notebook.ipynbProject-local, notebook, container, Kubernetes, Helm, and application configuration files
Kubernetes Secrets, Docker secrets, CI/CD variables, cloud secret managers, hosted app config vars, IDE/agent secrets, and password vaults
Python/TypeScript/Java/Go clients, notebooks, RAG apps, LangChain/LlamaIndex integrations, MCP servers, tests, examples, and committed deployment code
API debug logs, SDK traces, notebook output, CI logs, proxy logs, exception reports, and support bundles
Notes
Pinecone key values are opaque and no stable public prefix is documented. Scanner logic should prioritize env/header/client-argument context rather than value shape. PINECONE_TOKEN and host/environment variables were omitted from the scanner-facing environment location because the sourced docs emphasize PINECONE_API_KEY as the credential value; host, index, namespace, and project IDs are context rather than secrets.
Service Account OAuth Client Secret / Access Token
pinecone / service-account-oauth-secret
Pinecone Admin API service-account workflows use OAuth token creation at login.pinecone.io. Client credentials and resulting access tokens authorize Admin API requests such as project, API-key, and organization management.
Location
AuthorizationBearer access token used for Pinecone Admin API requests
https://login.pinecone.io/oauth/token, /oauth/tokenPinecone OAuth token endpoint contexts
PINECONE_CLIENT_SECRET, PINECONE_SERVICE_ACCOUNT_SECRET, PINECONE_ACCESS_TOKEN.env, .env.local, .env.production, pinecone.yml, pinecone.yaml, config.yaml, config.yml, docker-compose.yml, compose.yml, values.yaml, deployment.yamlAdmin automation, service-account, and deployment configuration files
Kubernetes Secrets, Docker secrets, CI/CD variables, cloud secret managers, admin automation vaults, and password vaults
admin automation scripts, Terraform/Pulumi code, CI workflows, tests, examples, and committed service-account clients
OAuth token exchange logs, Admin API traces, CI logs, proxy logs, and support bundles
Notes
Service-account secrets and access tokens are distinct from regular project API keys; both can be high-impact because they may create or revoke keys. Client and service-account IDs are not listed as credential-value locations; the client secret, service-account secret, and access token are the secret values.
Pinecone CLI / Target Context Credential
pinecone / cli-and-target-context
Pinecone CLI and IDE/agent integrations store target context and authenticate before selecting projects, indexes, or databases. Local config and agent workspace files may contain API keys, target hosts, project identifiers, or generated snippets from the console.
Location
PINECONE_API_KEYmcp.json, claude_desktop_config.json, cursor.json, settings.jsonIDE, MCP, and workspace configuration files that may contain Pinecone API keys or generated target snippets. Pinecone CLI target files can also exist in user config directories, but exact filenames were not confirmed.
mcp.json, claude_desktop_config.json, cursor.json, settings.jsonIDE, MCP, and workspace configuration files that may contain Pinecone API keys or generated target snippets. Pinecone CLI target files can also exist in user config directories, but exact filenames were not confirmed.
IDE secret stores, OS keychains, CI/CD variables, cloud secret managers, and password vaults
agent skill configs, MCP servers, AI coding tool snippets, notebooks, examples, and committed workspace files
CLI command output, IDE logs, agent logs, MCP traces, and CI output
Notes
The target context may not be secret by itself, but it often appears next to Pinecone API keys or service-account tokens.
Pinecone Assistant / Hosted Model API Key
pinecone / assistant-and-hosted-model-api-key
Pinecone Assistant and hosted inference APIs use Pinecone authentication to upload files, query assistants, generate embeddings, rerank, and manage assistant projects. Keys used in these contexts can expose assistant data, context snippets, and hosted-model usage.
Location
Api-Key, X-Pinecone-API-Key, AuthorizationPinecone Assistant, inference, embedding, and rerank API request contexts
PINECONE_API_KEY, PINECONE_ASSISTANT_API_KEY/assistant, /assistant/chat, /assistant/files, /embed, /rerank, /modelsPinecone Assistant and hosted inference endpoint contexts
.env, .env.local, .env.production, assistant.yaml, assistant.yml, pinecone.yml, pinecone.yaml, config.yaml, config.yml, notebook.ipynbAssistant, hosted model, notebook, and app configuration files
Kubernetes Secrets, Docker secrets, CI/CD variables, hosted app config vars, cloud secret managers, and password vaults
assistant clients, RAG apps, notebooks, examples, tests, frontend/backend code, and committed demos
assistant chat traces, file upload logs, inference request logs, notebook output, proxy logs, and support bundles
Notes
Pinecone Assistant can store files, snippets, and conversation context; API keys protecting it can expose more than vector index metadata.
Embedding / RAG Integration Provider API Key
pinecone / integration-provider-api-key
Pinecone applications commonly combine Pinecone keys with embedding, LLM, object-store, and data-source credentials. These are not Pinecone-native secrets, but Pinecone integrations often store them beside Pinecone config.
Location
OPENAI_API_KEY, ANTHROPIC_API_KEY, COHERE_API_KEY, VOYAGE_API_KEY, JINA_API_KEY, MISTRAL_API_KEY, GOOGLE_API_KEY, GEMINI_API_KEY, HUGGINGFACEHUB_API_TOKEN, HF_TOKEN, AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY.env, .env.local, .env.production, docker-compose.yml, compose.yml, values.yaml, deployment.yaml, config.yaml, config.yml, notebook.ipynbPinecone RAG, embedding, data-ingestion, and deployment configuration files
Kubernetes Secrets, Docker secrets, CI/CD variables, cloud secret managers, hosted app config vars, and password vaults
embedding pipelines, LangChain/LlamaIndex apps, notebooks, data ingestion jobs, tests, examples, and committed application code
embedding API errors, ingestion traces, notebook output, CI logs, and support bundles
Notes
Keep provider-specific token shapes in provider templates where available; this block captures the product-local storage context for Pinecone deployments.
Vector Index / Namespace / Import Data Leakage
pinecone / vector-index-and-namespace-data-leakage
Pinecone indexes, namespaces, records, metadata, imports, backups, and assistant context can contain proprietary content, PII, embedded secrets, retrieval context, or sensitive tenant/project identifiers.
Location
/vectors/upsert, /query, /vectors/fetch, /vectors/update, /vectors/delete, /describe_index_stats, /records/upsert, /namespaces, /bulk/importsPinecone data-plane and import endpoint contexts that can expose records, metadata, and namespace data
PINECONE_INDEX, PINECONE_NAMESPACE, PINECONE_PROJECT_ID, PINECONE_HOST, PINECONE_INDEX_HOST.env, .env.local, .env.production, import.yaml, import.yml, pinecone.yml, pinecone.yaml, config.yaml, config.yml, notebook.ipynbIndex, namespace, import, backup, and RAG application configuration files
vector IDs, records, metadata, namespace names, import records, assistant files, and retrieval context stored in Pinecone projects
vector export files, import manifests, backups, parquet/json/jsonl/csv datasets, notebook exports, support bundles, and model/RAG demo archives
query traces, retrieved document logs, import logs, embedding logs, application debug logs, notebook output, and backup/restore output
Notes
Treat this as context-dependent sensitive data. Scanner hits should identify places where Pinecone-connected apps expose data even when no credential value is present.
Scope
Authorized use
LOLCreds helps map the credential surface of real products: known defaults, generated values, credential locations, and exposure patterns.