syncing-aidlc-workflows

Pass

Audited by Gen Agent Trust Hub on Sep 27, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches the aidlc-workflows project from github.com/awslabs/aidlc-workflows. This is used to retrieve the latest tools and documentation for integration into the user's project. The skill supports both git cloning and direct tarball downloads from GitHub's codeload service.
  • [REMOTE_CODE_EXECUTION]: Executes a build script (node build/kiro-ide/build.js) on the content downloaded from the AWS Labs repository when the 'kiro' tool is selected. This is performed within a temporary directory to generate the necessary artifacts before they are integrated into the host project.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes markdown documentation from a remote source for translation. This creates a surface where instructions embedded in the external files could potentially influence the agent's behavior during the reading and translation process.
  • Ingestion points: Files downloaded from the awslabs/aidlc-workflows repository to the .aidlc-sync/base/ directory, which are subsequently read by the agent.
  • Boundary markers: None; the agent is instructed to read the raw markdown files from the base/ directory for translation.
  • Capability inventory: The skill has access to Bash, Write, Edit, Glob, Grep, and Read tools.
  • Sanitization: The skill does not perform content sanitization or filtering on the ingested markdown before the agent processes it.
  • [COMMAND_EXECUTION]: Utilizes several shell scripts (lib/*.sh) to automate complex tasks such as 3-way merging via git merge-file, hash calculation using sha256sum/shasum, and structured data processing with jq. It also includes a normalization step that strips sensitive fields from settings.json during the ingestion process.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 27, 2026, 11:59 PM