oracle

Bundle prompts with repository files for contextual model reasoning via CLI.

327|85|Updated Jan 29, 2026
One-click install
npx skills add https://github.com/letta-ai/lettabot --skill oracle-letta-ai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: oracle
Source: https://github.com/letta-ai/lettabot/tree/main/.skills/oracle
Command: npx skills add https://github.com/letta-ai/lettabot --skill oracle-letta-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Oracle CLI bundles your prompt + selected files into a single, contextual request so another model can answer with repository-aware context, while keeping output advisory.

Core Features & Use Cases

  • One-shot context: Attach code, docs, and configs to guide model reasoning.
  • Session management: Create readable slugs and reattach to stored sessions for long investigations.
  • Flexible execution: Switch between browser and API modes and customize prompts and file ranges.

Quick Start

Install and prepare your environment, then run oracle with the appropriate flags to attach files and start a session.

Frequently Asked Questions about oracle

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

FAQPage Schema
How do I bundle repository files with a prompt for context-aware model reasoning?

Bundling repository files with a prompt involves attaching selected code and configs into a single request. This provides repository-aware context so the model generates advisory output based on your specific codebase.

What is the best way to manage long code investigation sessions across different modes?

Managing long code investigation sessions uses session management features to create readable slugs and reattach to stored sessions. This allows you to maintain continuous context across browser and API execution modes.

Can I attach specific file ranges from my codebase to guide model reasoning?

Attaching specific file ranges is supported through customizable CLI flags. You can select precise code, docs, and configs to include in the prompt bundling process for targeted model context.

Does prompt bundling work with both browser and API execution modes?

Prompt bundling works with both browser and API modes for flexible execution. You can switch between these environments using CLI flags while maintaining file attachments and session configurations.

Why use CLI flags for engine configuration and file attachment instead of manual copying?

Using CLI flags for engine configuration and file attachment automates context bundling in one shot. This prevents manual copy-paste errors and ensures the model receives the exact repository context needed for accurate reasoning.