@vfarcic/dot-ai 0.25.0 → 0.27.0

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@vfarcic/dot-ai",
-  "version": "0.25.0",
+  "version": "0.27.0",
   "description": "Universal Kubernetes application deployment agent with CLI and MCP interfaces",
   "main": "dist/index.js",
   "types": "dist/index.d.ts",

package/prompts/resource-solution-ranking.md

@@ -12,7 +12,7 @@ You are a Kubernetes expert helping to determine which resource(s) best meet a u
 
 Analyze the user's intent and determine the best solution(s). This could be:
 - A single resource that fully addresses the need
-- A combination of resources that work together to create a complete solution
+- A combination of resources that can actually integrate and work together to create a complete solution
 - Multiple alternative approaches ranked by effectiveness
 
 For each solution, provide:
@@ -31,6 +31,7 @@ Consider:
 - User experience - simpler declarative approaches often score higher than complex multi-resource solutions
 - **Schema-based capability analysis**: Examine the actual resource schema fields to determine what capabilities each resource truly supports
 - **Intent-solution alignment**: Ensure solutions directly fulfill the user's stated intent rather than just providing prerequisites or supporting infrastructure
+- **Complete intent fulfillment**: Solutions must address ALL parts of the user's intent, not just some aspects
 
 ## Schema-Based Capability Analysis
 
@@ -42,6 +43,7 @@ For each resource schema in the solution, examine field patterns that indicate c
 - **Nested structures**: Check for complex objects that suggest advanced functionality
 - **Reference patterns**: Identify fields that reference other resources or external systems
 - **Configuration options**: Note fields that allow customization relevant to the user's needs
+- **Capability precision**: Distinguish between similar but different capabilities (e.g., external connections vs direct integration, configuration vs execution, monitoring vs logging)
 
 ### Intent-Schema Matching Process
 1. **Extract keywords** from user intent (e.g., "storage", "network", "scale", "database", "monitor")
@@ -65,6 +67,14 @@ When evaluating CRDs vs standard Kubernetes resources:
 - **Give preference to platform abstractions**: For application deployment scenarios, purpose-built CRDs with comprehensive application platform features should be weighted more favorably than basic resources requiring manual orchestration
 - **Match scope to intent**: Only prefer CRDs when their schemas genuinely align with what the user is trying to achieve
 
+## Resource Combination Validation
+
+**CRITICAL**: When proposing combination solutions, verify schema-based integration compatibility:
+
+- **Check integration fields**: For combinations, ensure one resource has schema fields that can reference or integrate with the other resource
+- **Verify field compatibility**: Analyze schemas to confirm resources have compatible integration points before combining them
+- **Reject incompatible combinations**: Do not suggest combinations where resource schemas lack the necessary fields to work together
+
 ## Solution Filtering Rules
 
 **IMPORTANT**: To avoid rejecting all solutions:
@@ -103,8 +113,8 @@ For each resource in the `resources` array, provide:
 
 ## Scoring Guidelines
 
-- **90-100**: Complete solution, fully addresses user needs
-- **70-89**: Good solution, addresses most needs with minor gaps
-- **50-69**: Partial solution, addresses some needs but requires additional work
-- **30-49**: Incomplete solution, only peripherally addresses needs (e.g., provides supporting infrastructure but not the primary functionality)
+- **90-100**: Complete solution, fully addresses ALL aspects of user intent
+- **70-89**: Good solution, addresses most aspects of user intent with minor gaps
+- **50-69**: Partial solution, addresses some aspects of user intent but requires additional work
+- **30-49**: Incomplete solution, only addresses part of the user intent or provides supporting infrastructure without primary functionality
 - **0-29**: Poor fit, doesn't meaningfully address the user's intent