# Capability-Gap Learning System

- id: capability-gap-learning-system
- type: prompt
- category: agent-governance
- detail: https://agents.proofandstate.com/products/capability-gap-learning-system

## Summary
Build a governed system that classifies why agent work is blocked, separates missing capability from missing permission, proposes the smallest justified improvement, validates it, and retries the original task without self-granting authority.

## Problem
Agents can misclassify permission or configuration failures as missing capabilities and expand their own authority instead of stopping or improving safely.

## Core outcomes
- Boundary classification
- Recurring-gap deduplication
- Governed capability proposals
- Separated implementation and activation
- Retry of the original task after validated activation

## Requirements
- Existing agent or task runtime
- Durable task and checkpoint storage
- Allowlisted tool or skill gateway
- Explicit approval records
- Registry for agents, skills, tools, plugins, workers, and policies
- Structured task results with evidence
- Test runner and synthetic fixtures
- Retry path for the original blocked task

## Boundaries
- Missing permission is not missing capability
- Blocked agents cannot approve their own expansion
- Implementation and activation are separate
- Forbidden work remains forbidden

## Tags
- agent-governance
- capability-gaps
- permissions
- toolgate
- validation

## Canonical source
- repository: AyobamiH/agent-shop-products (main)
- payload path: products/capability-gap-learning-system/PROMPT.md

## Not published
Price, ratings, reviews, sales figures and full PROMPT.md / SKILL.md payload bodies are not published.