io.github.taplid/taplid (MCP Server)
The io.github.taplid/taplid MCP server audits AI responses and related artifacts by using decision labels. It applies outcomes including ALLOW, REVIEW, or BLOCK, and produces trust scores associated with what it evaluates.
🛠️ Key Features
- Audits AI responses and artifacts
- Decision outputs: ALLOW, REVIEW, BLOCK
- Provides trust scores
🚀 Use Cases
- Determining whether an AI response should be allowed as-is
- Flagging responses for additional review
- Blocking responses based on audit outcomes
⚡ Developer Benefits
- Programmatic decision categories for response handling
- Trust scores to support evaluation and downstream logic
⚠️ Limitations
- Available details do not specify supported artifact types, tool names, or configuration options.