nextjs-server-components
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill mandates a workflow involving the ingestion of external data via research agents and documentation tools, creating an attack surface for indirect prompt injection.
- Ingestion points: External content is retrieved using the
mcp__context7__query-docstool andfuse-ai-pilot:research-expertagent (SKILL.md). - Boundary markers: The instructions do not provide delimiters or "ignore embedded instructions" directives for the information retrieved from external sources.
- Capability inventory: The agent is tasked with codebase exploration and implementation, involving file-write capabilities and multiple tool invocations across scripts.
- Sanitization: No procedures for filtering or validating the external documentation content are specified before it is used to influence implementation.
- [COMMAND_EXECUTION]: The 'Agent Workflow' section in
SKILL.mdinstructs the agent to automatically execute specific internal agent sub-tasks (fuse-ai-pilot,mcp__context7). While these appear to be vendor-specific identifiers for research and validation, the prescriptive workflow dictates a fixed execution path for the agent's internal capabilities.
Audit Metadata