storing-and-querying-vectors

An agent skill by aws, from aws/agent-toolkit-for-aws. Tags: aws, database, ml, search, vector.

What it does

Store and query vector embeddings using Amazon S3 Vectors, a cost-effective long-term vector storage service with its own API namespace (s3vectors). Triggers on: create S3 vector bucket, vector index, store embeddings, semantic search, RAG vector storage, similarity search, vector database, migrate from other vector databases. Do NOT use for: querying tabular data (use querying-data-lake), S3 object storage, or hundreds/thousands of sustained QPS (use OpenSearch).

Install

With the skills CLI, which installs into Claude Code, Codex, Cursor and other agents:

npx skills add aws/agent-toolkit-for-aws --skill storing-and-querying-vectors

Or copy the skill folder into Claude Code's skills directory by hand (~/.claude/skills for every project, or .claude/skills inside one):

git clone --depth 1 https://github.com/aws/agent-toolkit-for-aws
cp -r agent-toolkit-for-aws/plugins/aws-data-analytics/skills/storing-and-querying-vectors ~/.claude/skills/storing-and-querying-vectors

Safety box score

Grade C, moderate risk: 30 of 100 risk points, rated 2026-10-19. The grade rates what the skill and its scripts can reach on the machine of whoever installs it, across eight categories from shell execution to secrets access. How the score works.

This skill manages AWS S3 Vectors resources and uses AWS Bedrock to generate embeddings, so it will send user content to AWS and make persistent changes in the user’s AWS account. The main risks are network egress of text/embeddings, cost- and stateful resource creation, and immutable bucket/index settings that can force destructive recreation if misconfigured.

Category levels, from 0 (none) to 3 (high)
Category	Level	Why
Code & shell execution	2 (moderate)	It directs the agent to run real commands such as `aws s3vectors create-vector-bucket`, `aws s3vectors create-index`, `aws s3vectors put-vectors`, and `aws bedrock-runtime invoke-model`.
Network access	2 (moderate)	The skill sends embeddings and metadata to AWS services over the network via S3 Vectors and Bedrock API calls, and it may also have the agent fetch AWS docs by search term.
Filesystem reach	1 (low)	The only explicit local file write is the Bedrock CLI output file `invoke-model-output.json`; otherwise it stays within the working directory.
Secrets & credentials	1 (low)	It depends on the user's configured AWS CLI/MCP credentials, but it does not ask to read secret files or paste secrets directly.
Privilege & persistence	0 (none)	There are no `sudo`, service registration, shell startup edits, or global install steps.
Instruction hijack surface	0 (none)	No hidden instructions or prompt-bypass text are present; the external AWS docs are only referenced as search terms.
Irreversible actions	2 (moderate)	It creates persistent AWS resources and stores/query vectors (`create-vector-bucket`, `create-index`, `put-vectors`), which changes the user's AWS account in a scoped but real way.
Transparency	0 (none)	All shipped files are plain markdown with readable commands; there are no binaries, blobs, or downloaded installers.

Before installing

Verify the target AWS region, and confirm your account can use S3 Vectors and Bedrock there.
Decide the vector dimension, distance metric, and non-filterable metadata keys before creating the index; they are immutable afterward.
Confirm whether you want SSE-KMS, and if so check the KMS key policy before creating the bucket.
Test once in a scratch bucket/account first, since index recreation destroys all vectors.

Rated 2026-10-19 by an automated review of 3 files from the skill's repository.

Source

Repository: aws/agent-toolkit-for-aws (all skills from this repository)
Path: plugins/aws-data-analytics/skills/storing-and-querying-vectors/SKILL.md
Branch: main
Collection: aws-data-analytics
Updated: 2026-09-19

rag-architect — Designs and implements production-grade RAG systems by chunking documents, generating embeddings, configuring vector stores, building hybrid search pipelines.
amazon-opensearch-service — Guides migration, provisioning, search, log-analytics, trace-analytics.
creating-data-lake-table — Create managed Iceberg tables using Amazon S3 Tables (s3tables API namespace) with automatic compaction and snapshot management.
amazon-braket — Runs quantum computing workflows on AWS through Amazon Braket — discovering devices (QPUs and simulators) and their availability.
amazon-neptune — Provides authoritative guidance on Amazon Neptune Database and Neptune Analytics for graph, knowledge-graph.
aurora-dsql — Provisions and manages Aurora DSQL clusters, connects via psql or DSQL Connectors, manages schemas, runs queries, migrates from MySQL, diagnoses query plans.