frontend-review-state
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it ingests and processes untrusted data from the user's project files.
- Ingestion points: Reads
package.jsonand various source files in thesrc/directory usinggrepand file sampling. - Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded prompts in the analyzed source code.
- Capability inventory: The skill has the ability to read files and write a report file to
<client-repo>/.frontend-review/report/latest/md/state-review.md. - Sanitization: There is no evidence of sanitization or filtering of the content read from the source files before it is processed by the agent.
- [EXTERNAL_DOWNLOADS]: The documentation describes installation procedures that fetch content from the author's repositories.
- Evidence: The README mentions
npx degit mizchi/skills/frontend-review-stateandnpx skills add mizchi/skills. - Context: These references point to the skill author's own GitHub repositories and represent standard installation patterns for this vendor's tools.
Audit Metadata