prompt-engineering-patterns

Master advanced prompt engineering techniques to maximize LLM performance, reliability, and controllability in production. Use when optimizing prompts, improving LLM outputs, or designing production prompt templates.

1.16x

Quality

41%

Does it follow best practices?

Impact

85%

1.16x

Average score across 3 eval scenarios

Securityby

Passed

No known issues

Optimize this skill with Tessl

npx tessl skill review --optimize ./tests/ext_conformance/artifacts/agents-wshobson/llm-application-dev/skills/prompt-engineering-patterns/SKILL.md

Evaluation results

82%

Resume Information Extractor

LangChain structured output with Pydantic schemas

Criteria

Without context

With context

ChatAnthropic import

100%

Correct model name

with_structured_output usage

100%

Pydantic BaseModel schema

100%

Field descriptions

100%

Literal or constrained type

ChatPromptTemplate usage

100%

ValidationError fallback

50%

100%

Fallback confidence flag

100%

LangChain chain composition

100%

No hardcoded values in prompt

100%

79%

33%

Adaptive Customer Support Response Generator

Dynamic few-shot selection with VoyageAI and Chroma

Criteria

Without context

With context

VoyageAI embeddings import

50%

voyage-3-large model

66%

Chroma vector store

100%

SemanticSimilarityExampleSelector

100%

k=2 selection count

Example format: User/Assistant

100%

Different examples per query

100%

Consistent example formatting

100%

prompt_log.json output

100%

No hardcoded example selection

100%

96%

-2%

Prompt Quality Improvement for a Bug Triage System

Iterative prompt optimization with metrics and A/B testing

Criteria

Without context

With context

Baseline established first

100%

Progressive disclosure in variants

100%

Accuracy metric tracked

100%

Token count metric tracked

100%

Success rate metric tracked

100%

Variants compared systematically

100%

optimization_results.json exported

100%

Parameterized prompts

100%

prompt_design_notes.md

100%

Next step documented

100%

One-change-at-a-time principle

75%

50%

Winner identified with evidence

100%

Repository: Dicklesworthstone/pi_agent_rust
Commit: 47823e3

Evaluated: 29 days ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Resume Information Extractor Adaptive Customer Support Response Generator Prompt Quality Improvement for a Bug Triage System

Is this your skill?

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.