advanced-react
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest and process untrusted external data, specifically React source code,
package.jsonfiles, and lockfiles. This creates an attack surface for indirect prompt injection, where an attacker could embed malicious instructions within the source code or metadata of a project being analyzed. These instructions could potentially influence the agent's output during planning or review phases. The skill lacks explicit boundary markers or instructions to treat analyzed code as passive data, although it does enforce a structured output contract which mitigates some risks. - Ingestion points: Target repository code files,
package.json, and package lockfiles. - Boundary markers: None explicitly defined in the instructions to separate code analysis from instruction execution.
- Capability inventory: The skill allows for file reading, planning composition structures, generating code, and providing reviews. It does not appear to invoke external network tools or write sensitive system files.
- Sanitization: No explicit sanitization or filtering of the ingested source code is performed before analysis.
Audit Metadata