solid-react
Pass
Audited by Gen Agent Trust Hub on Feb 23, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill presents an attack surface for indirect prompt injection through its automated analysis and research workflow.
- Ingestion points: The agent is instructed to read external search results from Exa, documentation from Context7, and existing project codebase files (as mandated in SKILL.md).
- Boundary markers: Absent. The skill does not provide delimiters or "ignore instructions" warnings for data retrieved from external sources or local files.
- Capability inventory: The skill uses TeamCreate to spawn specialized sub-agents (fuse-ai-pilot) and research tools that interact with the internet.
- Sanitization: Absent. Content retrieved during research or codebase exploration is processed without validation or filtering.
- [COMMAND_EXECUTION]: The workflow involves invoking specialized internal tools and sub-agents for automation.
- Evidence: SKILL.md mandates the use of TeamCreate for spawning fuse-ai-pilot agents (explore-codebase, research-expert, sniper) and requires the use of MCP tools (mcp__context7__query-docs).
- Note: These tools are vendor-provided (fusengine) or part of the platform's standard Model Context Protocol (MCP) toolset.
- [EXTERNAL_DOWNLOADS]: The skill's code templates and implementation guides reference multiple third-party libraries and services.
- Evidence: Templates depend on established React ecosystem packages including @tanstack/react-query, @tanstack/react-router, zustand, zod, and date-fns.
- Evidence: The guidelines mention research via Exa and integration with analytics platforms like Google Analytics (gtag) and Mixpanel.
- Note: These are recognized, well-known, and trusted services/libraries within the web development community.
Audit Metadata