skills.rest
black-yt
black-ytCOMMUNITY
Explore all skills by this author →
Website𝕏@wanghan_xuGitHub

context-overlay

Configure and debug an OpenAI-compatible context overlay proxy with YAML rules.

7|1|Updated May 9, 2026
Search Tags:#streaming#prompt injection#proxy#openai-compatible#request routing#skill retrieval#rule matching
One-click install
npx skills add https://github.com/black-yt/skills --skill context-overlay
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: context-overlay
Source: https://github.com/black-yt/skills/tree/main/context-overlay
Command: npx skills add https://github.com/black-yt/skills --skill context-overlay

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you configure and debug an OpenAI-compatible context overlay proxy that can deterministically inject prompts, apply rules, and route requests without changing your client applications.

Core Features & Use Cases

  • Rule-based prompt injection: Add system, user, or patched content when specific paths, models, messages, or request fields match.
  • Request routing and rejection: Send selected requests to different upstreams or block unsafe requests with explicit rules.
  • Skill-dir retrieval and streaming forwarding: Load relevant skill JSON from a directory, forward OpenAI-compatible traffic, and validate behavior locally or through a tunnel.
  • Use Case: You need to add a consistent planning instruction to scientific chat requests while preserving tools, streaming, and multimodal fields.

Quick Start

Ask the AI to help you create or debug a context-overlay config for a local OpenAI-compatible proxy with rule-based prompt injection and request routing.

Frequently Asked Questions about context-overlay

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

FAQPage Schema
How do I inject system prompts into OpenAI-compatible API requests without changing my client application?▼

You can inject system prompts into OpenAI-compatible API requests by configuring a context overlay proxy that deterministically adds or patches content based on rule matching, leaving your client application untouched.

Can I route specific chat completion requests to different upstream models using a proxy?▼

Yes, you can route specific chat completion requests to different upstreams by configuring request routing rules within the context overlay proxy to selectively forward traffic based on matching paths, models, or request fields.

How does prompt injection work with streaming chat completions and multimodal fields?▼

Prompt injection with streaming chat completions preserves tools, image_url, response_format, and stream fields by applying rule-based context augmentation to OpenAI-compatible traffic without dropping or altering these existing multimodal request properties.

What is the best way to block unsafe prompts before forwarding them to an upstream LLM?▼

The best way to block unsafe prompts is by configuring explicit rejection rules in an OpenAI-compatible context overlay proxy, which evaluates request fields and messages to block unsafe requests before upstream forwarding occurs.

Do I need YAML configuration to set up deterministic prompt and context injection?▼

Yes, you need YAML configuration to set up deterministic prompt and context injection, as it defines the rule matching, request routing, and environment variable expansion required for safe authentication and local proxy validation.

Does the context overlay proxy support retrieving and loading skill directories dynamically?▼

Yes, the context overlay proxy supports skill directory retrieval by loading relevant skill JSON from a configured directory, allowing dynamic context augmentation and validation during local proxy operation and upstream forwarding.

Related Skills

Explore More Software Engineering Skills→
justinlevinedotmejustinlevinedotme
7

security-vite

Audit Vite apps for VITE_* exposure, secret leakage, and proxy misconfigurations.

Community
Advanced
polymerdaopolymerdao
1

polymer-pay-llm

Route OpenAI-compatible chat completions through Polymer Pay's category-based provider routing.

Official
Advanced
robabbyrobabby

Professional Profile Skill

Load professional profile files to personalize Claude Code coding assistance.

Community
Basic
photonics-dhlphotonics-dhl

tavily-proxy-workaround

Call Tavily API through explicit HTTP proxies using Python urllib.

Community
Intermediate