design-md-validator
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches the
@google/design.mdpackage from the npm registry usingnpx. It also allows for downloading remoteDESIGN.mdfiles from URLs provided by the user. - [REMOTE_CODE_EXECUTION]: Executes the
@google/design.mdpackage dynamically vianpx. This is the intended behavior for interacting with this tool and utilizes a package from a well-known organization. - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from
DESIGN.mdfiles, which could contain instructions designed to influence the agent's behavior when reviewing linting results. - Ingestion points: Processes local files and downloads content from remote URLs specified by the user (referenced in
SKILL.md). - Boundary markers: There are no explicit delimiters or instructions for the agent to ignore embedded commands within the files being linted.
- Capability inventory: The skill possesses the ability to execute shell commands and perform file operations via its defined workflows.
- Sanitization: No sanitization or validation of the markdown content is performed prior to the linter execution or agent review.
Audit Metadata