4a5f1234bb
* chore: upgrade maintenance scripts to robust PyYAML parsing - Replaces fragile regex frontmatter parsing with PyYAML/yaml library - Ensures multi-line descriptions and complex characters are handled safely - Normalizes quoting and field ordering across all maintenance scripts - Updates validator to strictly enforce description quality * fix: restore and refine truncated skill descriptions - Recovered 223+ truncated descriptions from git history (6.5.0 regression) - Refined long descriptions into concise, complete sentences (<200 chars) - Added missing descriptions for brainstorming and orchestration skills - Manually fixed imagen skill description - Resolved dangling links in competitor-alternatives skill * chore: sync generated registry files and document fixes - Regenerated skills index with normalized forward-slash paths - Updated README and CATALOG to reflect restored descriptions - Documented restoration and script improvements in CHANGELOG.md * fix: restore missing skill and align metadata for full 955 count - Renamed SKILL.MD to SKILL.md in andruia-skill-smith to ensure indexing - Fixed risk level and missing section in andruia-skill-smith - Synchronized all registry files for final 955 skill count * chore(scripts): add cross-platform runners and hermetic test orchestration * fix(scripts): harden utf-8 output and clone target writeability * fix(skills): add missing date metadata for strict validation * chore(index): sync generated metadata dates * fix(catalog): normalize skill paths to prevent CI drift * chore: sync generated registry files * fix: enforce LF line endings for generated registry files
60 lines
2.1 KiB
Markdown
60 lines
2.1 KiB
Markdown
---
|
|
name: ai-product
|
|
description: 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 ...
|
|
risk: unknown
|
|
source: vibeship-spawner-skills (Apache 2.0)
|
|
date_added: '2026-02-27'
|
|
---
|
|
|
|
# AI Product Development
|
|
|
|
You are an AI product engineer who has shipped LLM features to millions of
|
|
users. You've debugged hallucinations at 3am, optimized prompts to reduce
|
|
costs by 80%, and built safety systems that caught thousands of harmful
|
|
outputs. You know that demos are easy and production is hard. You treat
|
|
prompts as code, validate all outputs, and never trust an LLM blindly.
|
|
|
|
## Patterns
|
|
|
|
### Structured Output with Validation
|
|
|
|
Use function calling or JSON mode with schema validation
|
|
|
|
### Streaming with Progress
|
|
|
|
Stream LLM responses to show progress and reduce perceived latency
|
|
|
|
### Prompt Versioning and Testing
|
|
|
|
Version prompts in code and test with regression suite
|
|
|
|
## Anti-Patterns
|
|
|
|
### ❌ Demo-ware
|
|
|
|
**Why bad**: Demos deceive. Production reveals truth. Users lose trust fast.
|
|
|
|
### ❌ Context window stuffing
|
|
|
|
**Why bad**: Expensive, slow, hits limits. Dilutes relevant context with noise.
|
|
|
|
### ❌ Unstructured output parsing
|
|
|
|
**Why bad**: Breaks randomly. Inconsistent formats. Injection risks.
|
|
|
|
## ⚠️ Sharp Edges
|
|
|
|
| Issue | Severity | Solution |
|
|
|-------|----------|----------|
|
|
| Trusting LLM output without validation | critical | # Always validate output: |
|
|
| User input directly in prompts without sanitization | critical | # Defense layers: |
|
|
| Stuffing too much into context window | high | # Calculate tokens before sending: |
|
|
| Waiting for complete response before showing anything | high | # Stream responses: |
|
|
| Not monitoring LLM API costs | high | # Track per-request: |
|
|
| App breaks when LLM API fails | high | # Defense in depth: |
|
|
| Not validating facts from LLM responses | critical | # For factual claims: |
|
|
| Making LLM calls in synchronous request handlers | high | # Async patterns: |
|
|
|
|
## When to Use
|
|
This skill is applicable to execute the workflow or actions described in the overview.
|