Bundle repository files and prompts into a context payload for AI models.
npx skills add https://github.com/me2Doc/friendlyclaw --skill oracle-me2doc
Please help me install this Agent Skill. Skill: oracle Source: https://github.com/me2Doc/friendlyclaw/tree/main/body/skills/oracle Command: npx skills add https://github.com/me2Doc/friendlyclaw --skill oracle-me2doc
This skill solves the context-window limitation by intelligently bundling your prompt and relevant repository files into a single, high-signal payload for advanced AI models.
Use the oracle skill to analyze the current directory with a prompt asking for a summary of the project architecture.
High-intent search queries and answers about installing and using this skill.
To bundle repository context for AI code analysis, this skill aggregates your code, documentation, and configuration files with a prompt into a single high-signal payload. It respects .gitignore rules and estimates tokens to prepare a context-rich request for external AI models.
Context bundling helps debugging by packaging entrypoints, configuration files, and error logs together. Sending this aggregated payload to a high-reasoning AI model enables a deep-dive architectural review without losing critical file relationships.
Yes, you need the oracle CLI binary installed and appropriate environment configuration. This CLI interfaces with browser or API-based AI engines to process the bundled repository files and user prompts for external analysis.
Yes, you can detach and reattach long-running asynchronous analysis sessions. This session management capability allows you to start a complex code review and check back on the AI processing results later.
The best way to overcome context window limitations is intelligently bundling relevant repository files with your prompt. This method manages file selection and token estimation to ensure only high-signal context is sent to the AI model.
FatByBundle prompts with selected files and directories for LLM context.
mattstyles333Orchestrate a six-phase research process to generate multi-perspective analytical reports.
johnlindquistBundles code files into AI-ready context with a 49k token limit.
steipeteBundle prompts and files into a single one-shot request for repository-context aware answers.
