amazon-dynamodb

An agent skill by aws, from aws/agent-toolkit-for-aws. Tags: database, debugging, design, dynamodb, performance.

What it does

Designs, reviews, and debugs DynamoDB data layers from design axioms — enumerates access patterns, chooses partition/sort keys and GSIs, decides single-table vs. multi-table, configures Streams, Global Tables, TTL, vector indexes for similarity search, and zero-ETL integrations to OpenSearch/Redshift/SageMaker Lakehouse, and produces a defensible data-layer design with a monthly cost estimate and optional live validation. Applies whenever a user is designing, reviewing, or refactoring anything backed by DynamoDB — schemas, access patterns, GSIs, single- vs. multi-table choices, Streams consumers, transactional outboxes, Global Tables, zero-ETL pipelines, or storing embeddings and running semantic/vector similarity search with SearchVectors on items already in DynamoDB — even when they don't say "axioms" or "design review." Also applies when debugging hot partitions, throttling, unbounded Scans, LWW conflicts, or surprise bills on DynamoDB workloads.

Install

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

npx skills add aws/agent-toolkit-for-aws --skill amazon-dynamodb

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/skills/specialized-skills/database-skills/amazon-dynamodb ~/.claude/skills/amazon-dynamodb

Safety box score

Grade D, elevated risk: 56 of 100 risk points, rated 2026-10-20. 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 is a DynamoDB design-and-benchmark harness that can write local artifacts and, when explicitly used for live validation, create real DynamoDB tables, a Lambda function, and an IAM role in a user-supplied AWS account. The main risks are destructive cloud changes, real AWS charges, and handling of AWS account identity/credential metadata; local filesystem impact is mostly limited to the working directory plus skill-root discovery in the user’s home tree.

Category levels, from 0 (none) to 3 (high)
Category	Level	Why
Code & shell execution	3 (high)	It is built around runnable scripts and shell commands such as `python3 "$DDB_SKILL_DIR/scripts/deploy_model.py"`, `python3 "$DDB_SKILL_DIR/scripts/benchmark_model.py"`, and `bash teardown.sh --confirm`, so installing it grants a live code-execution workflow.
Network access	3 (high)	The skill explicitly talks to real AWS services using boto3 and the AWS CLI against the ambient credential chain: `the skill calls real AWS APIs when it validates a design, deploys a scratch table or benchmarks one` and the teardown script runs `aws dynamodb delete-table`, `aws lambda delete-function`, and `aws iam delete-role`.
Filesystem reach	2 (moderate)	It writes many artifacts into the working directory (`dynamodb_data_model.json`, `cost_report.md`, `benchmark_config.json`, `created_resources.json`, `performance_report.md`, `teardown.sh`) and its resolver searches user home directories like `$HOME/.claude`, `$HOME/.kiro`, `$HOME/.codex`, and `$HOME/.cursor`.
Secrets & credentials	2 (moderate)	The workflow consumes AWS credentials via the ambient profile/SSO chain and records account metadata in artifacts; for example, `benchmark_config.json` stores an `aws_profile`, and the teardown script embeds the manifest’s account/prefix/profile hint, while `created_resources.json` is described as containing the account ID and resource ARNs.
Privilege & persistence	0 (none)	There is no evidence of sudo use, global installs, service registration, shell-startup edits, or OS-level privilege changes on the local machine.
Instruction hijack surface	0 (none)	I did not find hidden prompts, fetched instructions, or attempts to override the agent’s rules; the text is workflow guidance and guardrails rather than prompt-injection content.
Irreversible actions	3 (high)	The skill’s purpose includes creating and later deleting real cloud resources, and its live-validation path requires explicit consent because it will provision tables/Lambda/IAM and can generate AWS charges.
Transparency	2 (moderate)	Several core implementation files were not readable in the supplied review (`scripts/deploy_model.py`, `scripts/benchmark_model.py`, `scripts/calculate_costs.py`, `scripts/iterate_design.py`, `scripts/benchmark_lambda.py`), so the actual AWS and benchmark behavior is only partially inspectable.

Before installing

Read `scripts/generate_teardown.py` and be comfortable with it deleting DynamoDB tables, a Lambda function, an IAM role, and optionally CloudWatch logs.
Use a sandbox AWS profile/region only; the live-validation path creates real AWS resources and the manifest records account and resource metadata.
Note that `scripts/find_skill_dir.sh` searches `~/.claude`, `~/.kiro`, `~/.codex`, `~/.cursor`, `~/.config`, and `~/.local/share` when locating the skill root.

Rated 2026-10-20 by an automated review of 6 files from the skill's repository.

Source

Repository: aws/agent-toolkit-for-aws (all skills from this repository)
Path: skills/specialized-skills/database-skills/amazon-dynamodb/SKILL.md
Branch: main
Updated: 2026-09-19

redshift-guide — Amazon Redshift is NOT PostgreSQL — corrects PostgreSQL-derived LLM mistakes; covers Redshift-specific SQL, DDL, COPY/UNLOAD, system views, metadata discovery.
database-optimizer — Optimizes database queries and improves performance across PostgreSQL and MySQL systems.
postgres-pro — Use when optimizing PostgreSQL queries, configuring replication, or implementing advanced database features.
sql-pro — Optimizes SQL queries, designs database schemas, and troubleshoots performance issues.
connecting-to-data-source — Create and troubleshoot AWS Glue connections to JDBC databases (Oracle, SQL Server, PostgreSQL, MySQL, RDS), Redshift, Snowflake, and BigQuery.
app-platform-troubleshooting — Debug running App Platform applications by accessing containers, analyzing logs, running diagnostics, and applying fixes.