Orchestrate multi-session research audits across LLM vendors with structured contracts.
npx skills add https://github.com/Ronkupper/PRISM --skill core-ronkupper
Please help me install this Agent Skill. Skill: core Source: https://github.com/Ronkupper/PRISM/tree/main/plugins/prism/skills/core Command: npx skills add https://github.com/Ronkupper/PRISM --skill core-ronkupper
PRISM solves the challenge of conducting rigorous, multi-session research and audits by providing a structured framework that orchestrates tasks across multiple LLM vendors, ensuring consistency, completeness, and convergence.
Begin a PRISM audit by attaching your subject brief and the PRISM core skill to a new Claude session and asking the framework to initialize the setup probes.
High-intent search queries and answers about installing and using this skill.
Multi-vendor LLM research is orchestrated by dispatching atomic prompts to specialized models like Claude, ChatGPT, and Gemini. The framework synthesizes these outputs, ensuring consistency and convergence across complex audit workflows.
Maintaining state in multi-session LLM audits requires a continuous Master document and a What's next artifact. This living framework tracks engagement state, allowing seamless resumption of complex research tasks across different sessions.
Context-pressure telemetry in LLM orchestration frameworks manages complex engagement state by monitoring context limits. This telemetric framework ensures optimal prompt dispatch and prevents context overload during multi-vendor research.
Silent-omission risks in LLM research are prevented using a rigorous audit framework. It implements a triple-contract execution model and a canonical Lens Library to identify gaps, enforce audit standards, and grade coverage.
Complex research workflows can be resumed after interruption by leveraging continuous state management. The framework maintains a living Master document that tracks engagement state, ensuring seamless continuation across new sessions.
The triple-contract execution model enforces rigorous audit standards via Envelope, Self-check, and Output contracts. While it mitigates silent-omission risks, this strict structure requires strict compliance, potentially increasing setup overhead for simple research tasks.
vetacreativalaserLoad CLAUDE_PROMPT.md instructions into Claude's session context.
RonkupperOrchestrate multi-session research audits across multiple LLM vendors.
apassuelloEvaluate LLM outputs with automated metrics, human judgments, and benchmarking.
DharambirAgrawalRun one-shot Q&A, summaries, and generation via the Gemini CLI.
