prestashop-pr-qa
Warn
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: MEDIUMDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [DYNAMIC_EXECUTION]: The runner scripts (run.js, run-cli.js, run-http.js) dynamically load the scenario.js file using the require() function from a computed path based on command-line arguments. This file contains executable JavaScript code generated at runtime.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from GitHub pull requests and issues using gh pr view and gh issue view. This data is processed by the agent to author the executable scenario.js file, creating a risk that malicious instructions in a PR could influence the generated code.
- Ingestion points: gh pr view and gh issue view in SKILL.md.
- Boundary markers: None specified for the external ticket content.
- Capability inventory: Shell access, file writing, and dynamic JavaScript execution.
- Sanitization: None; the agent relies on its own instruction following to extract steps.
- [COMMAND_EXECUTION]: The run-cli.js script provides a shell execution capability via execFile('/bin/sh', ['-c', command], ...), allowing for the execution of arbitrary system commands as defined in the generated test scenario.
- [EXTERNAL_DOWNLOADS]: The scripts/playwright-lab.sh script downloads and installs the playwright package and browser binaries (chromium, ffmpeg) from the official NPM registry at runtime. This is performed to set up a transient testing environment without polluting the project repository.
Audit Metadata