artifacts-builder
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill requires the agent to execute local bash scripts (
scripts/init-artifact.shandscripts/bundle-artifact.sh) to perform project initialization and asset bundling. These scripts interact with the local filesystem and the Node.js package manager to install development dependencies. - [INDIRECT_PROMPT_INJECTION]: The skill facilitates a workflow where the agent processes and bundles external code. This creates an indirect prompt injection surface.
- Ingestion points: Files within the generated frontend project created in
Step 1and modified inStep 2. - Boundary markers: None provided in the skill instructions to separate project code from agent instructions.
- Capability inventory: The skill uses shell script execution, Node.js package installation, Parcel bundling, and potentially browser automation tools like Playwright or Puppeteer for testing.
- Sanitization: No specific sanitization or validation of the artifact source code is mentioned before the bundling or testing phases.
Audit Metadata