AI and machine learning workflow covering LLM application development, RAG implementation, agent architecture, ML pipelines, and AI-powered features.
57
Quality
37%
Does it follow best practices?
Impact
94%
1.08xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/ai-ml/SKILL.mdRAG pipeline design workflow
Data pipeline design
100%
100%
Embedding model selection
100%
100%
Vector database selection
100%
100%
Chunking strategy
100%
100%
Reranking step
100%
100%
Caching implementation
100%
100%
Retrieval accuracy testing
100%
100%
Vector search optimization
100%
100%
Implementation ordering
100%
87%
Embeddings generation step
100%
100%
Hybrid search consideration
37%
100%
Similarity search pattern
100%
100%
Without context: $0.4220 · 2m 39s · 16 turns · 21 in / 7,953 out tokens
With context: $0.5661 · 2m 55s · 24 turns · 27 in / 9,068 out tokens
AI agent development workflow
Distinct agent roles
100%
100%
Role responsibilities
100%
100%
Tool integration
100%
100%
Memory system
100%
100%
Human-in-the-loop
100%
100%
Orchestration mechanism
100%
100%
Error handling
100%
100%
Framework recommendation
0%
100%
Agent architecture design
100%
100%
Orchestration tested
0%
0%
Inter-agent communication
100%
100%
Without context: $0.3945 · 2m 41s · 13 turns · 19 in / 8,778 out tokens
With context: $0.6965 · 3m 39s · 26 turns · 66 in / 12,073 out tokens
LLM integration and observability checklist
Streaming support
0%
0%
Error handling
100%
100%
API key security
100%
100%
Rate limiting
100%
100%
Token usage tracking
100%
100%
Tracing setup
100%
100%
Logging configuration
100%
100%
Evaluation/quality monitoring
25%
100%
Performance monitoring
100%
100%
Cost tracking
100%
100%
Alerting
100%
100%
Production checklist completeness
100%
100%
Without context: $0.5686 · 3m 12s · 25 turns · 78 in / 10,101 out tokens
With context: $0.6256 · 2m 51s · 24 turns · 379 in / 10,437 out tokens
8692fbc
Table of Contents
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.