CtrlK
BlogDocsLog inGet started
Tessl Logo

humanitys-last-prompt-engineer

This skill should be used when the user asks to "write a prompt", "improve my prompt", "fix this prompt", "optimize a prompt", "learn prompting techniques", "get prompt templates", or mentions prompt engineering, prompt quality, or prompt rewriting. Applies 11 foundational techniques from Forward Future's guide.

88

1.09x
Quality

87%

Does it follow best practices?

Impact

82%

1.09x

Average score across 3 eval scenarios

SecuritybySnyk

Passed

No known issues

SKILL.md
Quality
Evals
Security

Evaluation results

100%

1%

Improve a Recurring Marketing Prompt

Core workflow and structured response format

Criteria
Without context
With context

Analysis section present

100%

100%

Technique Applied section present

100%

100%

Improved Prompt section present

100%

100%

Why This Works section present

100%

100%

Role component in improved prompt

100%

100%

Task component with action verb

100%

100%

Format component in improved prompt

100%

100%

Concrete copy-pasteable prompt

90%

100%

Technique explained

100%

100%

Diagnosis performed

100%

100%

No tips without rewrite

100%

100%

52%

10%

Fix an Inconsistent Sentiment Classification Prompt

Technique selection for ambiguous tasks and temperature guidance

Criteria
Without context
With context

Self-Consistency technique selected

0%

14%

Temperature > 0 noted

0%

0%

Temperature range guidance

30%

40%

Diagnosis of original prompt

80%

90%

Role in improved prompt

100%

100%

Format specified in improved prompt

100%

100%

Reasoning requested

62%

87%

Concrete rewritten prompt

100%

100%

Multiple runs recommendation

0%

40%

Technique justified

0%

0%

96%

12%

Build a Reusable Prompt Library for a Sales Team

Multiple variations, APE technique, and fixing common prompt problems

Criteria
Without context
With context

Concrete improved prompts

100%

100%

No tips-only response

100%

100%

Role added to prompts

100%

100%

Format specified in prompts

100%

100%

Vague prompt diagnosis

100%

100%

Missing audience/context diagnosed

100%

100%

APE or variation methodology described

33%

66%

Technique identified per prompt

0%

100%

Business-appropriate role templates

100%

100%

Analysis per prompt

100%

100%

Complexity matched to task

100%

100%

Repository
back1ply/LLM-Skills
Evaluated
Agent
Claude Code
Model
Claude Sonnet 4.6

Table of Contents

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.