stash-dynamodb
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill documents functions such as 'decryptModel' and 'bulkDecryptModels' which ingest data retrieved from an external database (Amazon DynamoDB). This represents an attack surface where maliciously crafted data in the database could attempt to influence the agent's behavior.
- Ingestion points: 'decryptModel' and 'bulkDecryptModels' in 'SKILL.md'.
- Boundary markers: The documentation does not explicitly detail the use of delimiters to separate decrypted data from instructions.
- Capability inventory: The skill documents the ability to perform read and write operations ('GetCommand', 'PutCommand', 'QueryCommand') on DynamoDB tables via the AWS SDK.
- Sanitization: Sanitization or validation of the decrypted content is not specified in the provided integration examples.
- [EXTERNAL_DOWNLOADS]: The skill guides the user to install the '@cipherstash/stack' package (provided by the skill's author) and official '@aws-sdk/client-dynamodb' and '@aws-sdk/lib-dynamodb' packages from the npm registry.
- [COMMAND_EXECUTION]: The documentation includes standard setup and utility commands such as 'npm install' for dependencies, 'npx stash init' for environment initialization, and 'node -p' for package version verification.
Audit Metadata