The self-documenting graph behind Ghost Intelligence. Every capability is automatically discoverable, searchable, teachable, documented, linked and visualised.
Self-grading surfaces weak spots before they fail.
For users
The system grades its own subsystems. Scores planning, execution, research, memory and outcomes; forecasts degradation. Open the Self view to read the latest grades.
For developers
Self-Assessment (id: self-assessment) is a working AI capability rendered at /pulse/self. It scores planning, execution, research, memory and outcomes; forecasts degradation. It builds on: meta-cognition.
For investors
Self-Assessment: Self-grading surfaces weak spots before they fail. It is working at 79% confidence, demonstrating real, defensible capability rather than a slide.
Learning objectives
Explain what Self-Assessment does in one sentence.
Describe why Self-Assessment matters to the system.
Open /pulse/self and identify the live values it produces.
Trace how Self-Assessment relies on meta-cognition.
Examples
Open the Self view to read the latest grades.
Known limitations
Not linked from primary navigation; specialist surface.
Future improvements
Open the Self view to read the latest grades.
Raise completion to Production Ready with more live coverage.