vrt-check
Pass
Audited by Gen Agent Trust Hub on Aug 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
BashandTasktools to execute commands discovered within the local repository (e.g.,just,npmscripts,playwright,storybook-test-runner). It dynamically assembles commands based on the contents ofpackage.json,justfile, andMakefile. - [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection. It ingests untrusted data from the repository's files to decide its next actions and shell commands.
- Ingestion points: Reads project configuration files (
package.json,justfile,Makefile,chromatic.config.json,.loki), VRT tool outputs/reports, and version control history (git diff,git log). - Boundary markers: The instructions do not define delimiters or specific warnings to ignore embedded instructions within the ingested data.
- Capability inventory: The skill has access to
Bash,Task, andRead, allowing it to execute arbitrary shell commands and read local files. - Sanitization: There is no evidence of sanitization or validation of the content read from these external sources before it is used to influence the agent's logic or command construction.
Audit Metadata