Bundle natural language prompts with project files into a single AI input.
npx skills add https://github.com/kolbick/Kolb-Bot --skill oracle-kolbick
Please help me install this Agent Skill. Skill: oracle Source: https://github.com/kolbick/Kolb-Bot/tree/main/skills/oracle Command: npx skills add https://github.com/kolbick/Kolb-Bot --skill oracle-kolbick
This Skill streamlines the process of providing AI models with relevant project context by bundling prompts with selected files, enabling more accurate and context-aware responses.
oracle CLI, including file selection, engine choice, and session management for long-running tasks.Use the oracle CLI to preview the token count for a task by running oracle --dry-run summary -p "<task>" --file "src/**".
High-intent search queries and answers about installing and using this skill.
To bundle project files and prompts for AI context, combine natural language instructions with specified files and directories into a single input. This allows AI models to analyze code with full project awareness for debugging or content generation.
Yes, you can use glob patterns to include or exclude files when bundling code context. This file selection mechanism filters source files and directories, ensuring only relevant project components are passed to the AI model.
The oracle CLI works with both API and browser-based AI execution engines. It supports session management across these engines, allowing long-running tasks and contextual interactions within your software development workflow.
The best way to preview token count before sending files to an LLM is using a dry run. Execute the CLI with a dry-run flag, your prompt, and file globs to estimate the token size of the bundled request.
Prompt templating for code analysis works by structuring natural language prompts with file references into a unified input. This allows repeatable, formatted instructions to be bundled with project directories for consistent AI interactions.
You should not use a bundled one-shot AI request for debugging when your codebase exceeds token limits or requires multi-turn conversational adjustments. In these cases, granular file selection or session-based interactions are more effective.
FlexasaurusrexBundle prompts with files and manage sessions using the oracle CLI.
sentientspriteBundle natural language prompts with files and directories for AI context.
vut08905Bundle prompts with files and manage engines and sessions via the oracle CLI.
nimrodfisherPackage analysis task details and context sources into reusable context bundles.