Orchestrate multi-LLM discovery pipelines to plan and validate ideas.
npx skills add https://github.com/jdonohoo/vern-bot --skill discovery-jdonohoo
Please help me install this Agent Skill. Skill: discovery Source: https://github.com/jdonohoo/vern-bot/tree/main/skills/discovery Command: npx skills add https://github.com/jdonohoo/vern-bot --skill discovery-jdonohoo
This Skill orchestrates a repeatable, multi-LLM discovery pipeline that turns informal prompts and supporting files into a battle-tested, consolidated plan and actionable artifacts, removing the manual work of coordinating models and steps.
Start a discovery run for the idea "smart-calendar" using the default pipeline and save results to the current directory.
High-intent search queries and answers about installing and using this skill.
Multi-LLm discovery pipeline orchestration validates startup ideas by executing configurable analysis, refinement, chaos check, and consolidation sequences across models like Codex, Claude, and Gemini.
A multi-LLM discovery pipeline orchestrates models like Codex, Claude, and Gemini through sequential analysis, refinement, chaos checks, consolidation, and architect breakdown steps to validate ideas.
Yes, you can run idea validation across Claude, Codex, and Gemini using configurable pipelines with fallback strategies to ensure safe, repeatable multi-LLM execution.
To execute a multi-LLM pipeline from the command line, locate the plugin root and run the packaged CLI with explicit flags, enforcing platform-aware execution and safeguards against user-supplied paths.
Yes, the discovery pipeline supports resume-on-failure and max-retries to ensure safe, repeatable execution when orchestrating multi-LLM workflows.
Outputs from a multi-LLM idea validation pipeline include structured output files, pipeline-status summaries, optional persona reviews, and an Oracle vision summary for informed decision-making.
jdonohooCreate a standardized discovery folder structure with input and output directories.
dlantsConvert vague requirements into structured implementation plans saved to plans/<planName>.md.
ticruz38Schedule and manage local reminders in a self-contained SQLite store.
Shallow-WOrchestrate research idea discovery from direction to pilot-tested experiment plans.
