oracle

Bundle prompts and selected files into a single context-rich request.

Updated Jan 1, 2026
One-click install
npx skills add https://github.com/hcnimi/clawdis --skill oracle-hcnimi
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: oracle
Source: https://github.com/hcnimi/clawdis/tree/main/skills/oracle
Command: npx skills add https://github.com/hcnimi/clawdis --skill oracle-hcnimi

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Oracle bundles your prompt + selected files into one context-rich request so another model can answer with project context (API or browser automation). Treat output as advisory: verify against code + tests.

Core Features & Use Cases

  • Bundle prompts + selected files into a single request so the model sees repository context.
  • Attach files via flexible --file patterns to seed code, docs, and assets.
  • Supports engine and model selection and session management for long-running, contextual tasks.

Quick Start

Run oracle with a task and attach relevant files using --file patterns to seed the prompt.

Frequently Asked Questions about oracle

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

FAQPage Schema
How do I bundle prompts with files for richer AI context?

To bundle prompts with files, you attach selected files via --file patterns to seed code and docs into a single context-rich request. This ensures the model sees repository context when generating guidance.

How does prompt bundling work for code review sessions?

Prompt bundling for code review works by combining your prompt and selected code files into one request. Session management coordinates these prompt payloads with file attachments for long-running, contextual tasks.

Can I select a specific engine and model for prompt bundling?

Yes, you can select a specific engine and model for prompt bundling. The tool supports engine and model selection, allowing you to direct your context-rich request to the desired model.

What is the best way to include repository context in AI requests?

The best way to include repository context is bundling prompts and files into a single request. You attach files using flexible --file patterns to seed code, docs, and assets for the model.

Can I test prompt bundling without sending requests to the model?

Yes, you can test prompt bundling without sending requests by using the --dry-run flag. This allows you to verify the bundled prompt and file attachments before the actual model guidance is generated.

Are AI responses from bundled prompts safe to apply directly to code?

No, AI responses from bundled prompts are not safe to apply directly. You should treat the output as advisory and always verify the model guidance against your actual code and tests.