skills.rest
machenjie
machenjieCOMMUNITY
Explore all skills by this author →
GitHub

ai-product-extension

Enforce engineering discipline for LLM, RAG, and agentic product changes.

4|Updated May 16, 2026
Search Tags:#llm evaluation#ai safety#model governance#rag security#agent tool use#hallucination risk#mlops rollback
One-click install
npx skills add https://github.com/machenjie/rd-skills --skill ai-product-extension
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ai-product-extension
Source: https://github.com/machenjie/rd-skills/tree/main/src/domain-extensions/ai-product-extension
Command: npx skills add https://github.com/machenjie/rd-skills --skill ai-product-extension

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Prevents unsafe, leaking, or low-quality AI behavior by enforcing product-grade engineering rules for LLMs, RAG, agent tool use, evaluation, and production governance.

Core Features & Use Cases

  • LLM safety and trust-boundary controls: prompts are injection-aware, outputs are treated as probabilistic, and downstream use validates untrusted model output.
  • Permission-aware RAG requirements: retrieval must apply the same ACL/RBAC filters as the source system to avoid cross-tenant leakage.
  • Evaluation-driven release gates: requires ground-truth datasets and defines quality thresholds for hallucination, refusal correctness, and adversarial cases.
  • MLOps governance for AI lifecycle: enforces model registry versioning, training-serving skew checks, drift monitoring, and rollback readiness.

Quick Start

Apply the ai-product-extension skill to your proposed LLM/RAG/agent change and ask the agent to produce an AI risk assessment plus a release checklist that includes evaluation coverage, permission-filtering requirements, and required safety controls.

Frequently Asked Questions about ai-product-extension

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I prevent permission bypass and cross-tenant data leakage in RAG retrieval?▼

To prevent cross-tenant data leakage in RAG retrieval, you must apply the same ACL and RBAC permission filters as the source system during vector database assembly to ensure strict access control.

What is the best way to evaluate LLM hallucination risk before a production rollout?▼

Evaluating LLM hallucination risk requires using ground-truth datasets to define measurable quality thresholds for refusal correctness and adversarial cases, creating evaluation-driven release gates for safe deployment.

How do I secure agent tool use and function calling against prompt injection?▼

Securing agent tool use requires prompt-injection resistant assembly, enforcing strict tool allowlists for function calling, and validating untrusted model output downstream to mitigate execution risks.

What MLOps governance do I need for model lifecycle modifications and rollback readiness?▼

MLOps governance for model lifecycle modifications requires enforcing model registry versioning, continuous drift monitoring, training-serving skew checks, and automated rollback readiness protocols.

Can I apply AI safety rules to classical ML rollout and prompt template changes?▼

Yes, AI safety rules apply to both classical ML rollout and prompt template changes by enforcing product-grade engineering discipline, untrusted-output validation, and measurable safety thresholds.

Related Skills

Explore More Product & Management Skills→
Alex1980AlexAlex1980Alex

evaluation-benchmark

Quantify search and RAG output quality using standardized retrieval metrics.

Community
Advanced
Ethical-AI-SyndicateEthical-AI-Syndicate
1

model-lifecycle-manager

Manage AI/ML model registration, versioning, monitoring, drift detection, and retraining triggers.

Official
Advanced
robotijnrobotijn
3

feature-store-validator

Audit feature store configurations for training-serving skew and data leakage.

Community
Advanced
JaiminVadadoriyaJaiminVadadoriya
1

ai-system-testing

Test AI features including LLM prompts, tool calls, and nondeterministic outputs.

Community
Advanced