positron-demo-video
Pass
Audited by Gen Agent Trust Hub on Mar 29, 2026
Risk Level: SAFECOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill runs shell commands for repository research (
git log), test execution (npx playwright test), and video processing (ffmpeg,ffprobe). - [REMOTE_CODE_EXECUTION]: The skill creates Playwright test files based on a narrative and executes them to automate the demo recording. This local code generation and execution is central to the skill's functionality.
- [PROMPT_INJECTION]: The skill's research phase ingests untrusted data from the repository that could potentially influence agent behavior.
- Ingestion points: git logs, changed files, and existing e2e tests.
- Boundary markers: No delimiters or instructions are used to isolate untrusted data during ingestion.
- Capability inventory: The skill can execute shell commands and run generated test code.
- Sanitization: No sanitization is applied to data read from the repository before use.
Audit Metadata