iblai-api-agent-mcp
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides setup instructions involving the vendor's official GitHub repository (https://github.com/iblai/iblai-mcp) to access a catalog of pre-configured MCP servers.\n- [INDIRECT_PROMPT_INJECTION]: The skill manages tool integrations that process data from external MCP servers, representing an attack surface for indirect prompt injection via tool responses.\n
- Ingestion points: External tool data is ingested into the agent context from configured MCP server responses (SKILL.md).\n
- Boundary markers: The skill relies on platform-managed tool execution frames rather than explicit instructions-level delimiters.\n
- Capability inventory: The skill utilizes curl to perform management operations against vendor API endpoints (api.iblai.app), including server registration and OAuth connection lifecycle management (SKILL.md).\n
- Sanitization: The skill features a clean_output setting (default true) which strips HTML from external server outputs, providing a layer of protection against malicious or malformed content.\n- [COMMAND_EXECUTION]: The skill provides examples of curl commands to interact with the ibl.ai API. These commands are used for administrative tasks like server registration and credential binding within the vendor's ecosystem.
Audit Metadata