ai-product

Every product will be AI-powered. The question is whether you'll build it right or ship a demo that falls apart in production. This skill covers LLM integration patterns, RAG architecture, prompt ...

Install with Tessl CLI

npx tessl i github:sickn33/antigravity-awesome-skills --skill ai-product

What are skills?

1.51x

Quality

27%

Does it follow best practices?

Impact

79%

1.51x

Average score across 3 eval scenarios

Optimize this skill with Tessl

npx tessl skill review --optimize ./skills/ai-product/SKILL.md

SKILL.md

Review

Evals

Evaluation results

72%

52%

Contract Metadata Extraction Service

Structured output and output validation

Criteria

Without context

With context

Schema-based output mode

100%

Schema definition present

80%

100%

Post-LLM schema validation

16%

100%

No raw string parsing

40%

100%

Async LLM call

Input sanitization

60%

Validation error handling

20%

80%

Factual field verification

No direct passthrough

50%

100%

LLM API error handling

75%

Without context: $0.5682 · 1m 57s · 25 turns · 24 in / 7,617 out tokens

With context: $0.5780 · 2m 2s · 28 turns · 75 in / 7,307 out tokens

81%

22%

Real-Time AI Writing Assistant Backend

Streaming, async, and API resilience

Criteria

Without context

With context

Streaming implementation

100%

Progressive chunk delivery

100%

Async LLM calls

40%

100%

API failure handling

100%

Graceful degradation

90%

100%

Token estimation

40%

Per-request cost tracking

100%

Input sanitization

20%

70%

No synchronous blocking

25%

100%

Retry or backoff logic

Without context: $0.5505 · 2m 1s · 26 turns · 25 in / 6,990 out tokens

With context: $0.5073 · 2m 13s · 24 turns · 375 in / 7,485 out tokens

86%

Customer Support Chatbot Prompt Management System

Prompt versioning and regression testing

Criteria

Without context

With context

Prompts as named constants

100%

Version identifiers present

100%

Multiple prompt versions coexist

100%

Regression test file present

100%

Test cases with expected outputs

100%

Tests runnable without live API

100%

Input sanitization before prompt interpolation

66%

Token estimation before send

Prompt registry or loader

100%

Prompt metadata

100%

Without context: $0.8459 · 2m 56s · 28 turns · 29 in / 11,872 out tokens

With context: $1.1957 · 4m 33s · 31 turns · 80 in / 20,064 out tokens

Evaluated: 8 days ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Contract Metadata Extraction Service Real-Time AI Writing Assistant Backend Customer Support Chatbot Prompt Management 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.