Coordinate parallel prompts across Codex, Gemini, and Claude with retry logic.
npx skills add https://github.com/choru-k/skills-for-claude --skill call-ai
Please help me install this Agent Skill. Skill: call-ai Source: https://github.com/choru-k/skills-for-claude/tree/main/plugins/call-ai/skills/call-ai Command: npx skills add https://github.com/choru-k/skills-for-claude --skill call-ai
Coordinates parallel prompts across Codex, Gemini, and Claude to obtain diverse AI responses without manual coordination, streamlining multi-model evaluation.
Provide a single prompt to run across Codex, Gemini, and Claude in parallel and review the aggregated results.
High-intent search queries and answers about installing and using this skill.
Running parallel prompts across Claude, Codex, and Gemini requires a unified wrapper that dispatches a single prompt to all configured models simultaneously, obtaining diverse AI responses without manual coordination.
Comparing multiple AI responses without manual coordination is achieved through parallel multi-model execution that dispatches prompts to different AIs simultaneously and logs structured results to a .responses directory for auditing.
Yes, you need jq and bc installed as dependencies to coordinate parallel multi-model execution, as the unified wrapper relies on these utilities for handling structured data and calculations during prompt orchestration.
Yes, you can use Zellij pane visualization for real-time streaming of AI responses when the environment supports it, allowing you to monitor parallel prompt execution across multiple models as it happens.
Retry logic handles timeouts and errors automatically during parallel execution by re-attempting failed prompts per-model, ensuring robust multi-model orchestration even when individual AI services experience intermittent connectivity issues.
Per-model variants allow configurable selection between thorough and fast prompt execution modes for each AI model, enabling tailored parallel multi-model evaluation based on your specific latency and depth requirements.
EvolutionAPIRun the same prompt through Claude, Codex, and Gemini in parallel and synthesize results.
vasilyu1983Optimize LLM inference performance and cost with vLLM and quantization.
tianhuilLaunch and monitor dev servers in Zellij panes with OpenCode output.
moonlight-lupinDistribute prompts to multiple AI models and log blind comparison votes.
