quickjs-sandbox
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references and fetches documentation from https://sebastianwessel.github.io/quickjs/ and recommends the installation of the @sebastianwessel/quickjs package suite from the public NPM registry in SKILL.md.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an indirect prompt injection attack surface by design to perform its primary function. 1. Ingestion points: The skill fetches external documentation from the sebastianwessel.github.io and jsr.io domains in SKILL.md. 2. Boundary markers: The skill provides a structured prompt template for the research sub-agent including Goal, Context, and Specific Questions fields to constrain the agent's interpretation of external data. 3. Capability inventory: The skill utilizes the Task tool and WebFetch to retrieve and process remote documentation. 4. Sanitization: No explicit sanitization or filtering of the retrieved web content is documented in the skill instructions.
Audit Metadata