ai-pi-nv-kimi

Invoke Kimi K2.5 via NVIDIA Pi to assess targets and output JSON findings.

Updated Apr 26, 2026
One-click install
npx skills add https://github.com/EndUser123/cc-marketplace --skill ai-pi-nv-kimi
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ai-pi-nv-kimi
Source: https://github.com/EndUser123/cc-marketplace/tree/main/plugins/cc-skills-ai-cli/skills/ai-pi-nv-kimi
Command: npx skills add https://github.com/EndUser123/cc-marketplace --skill ai-pi-nv-kimi

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Quickly identify vulnerabilities and weaknesses in targets by orchestrating an adversarial evaluation with Pi and Kimi K2.5, producing structured findings that can guide remediation.

Core Features & Use Cases

  • Orchestrates adversarial testing by invoking Kimi K2.5 through Pi to probe targets.
  • Outputs a structured JSON score, one-sentence summary, and list of issues for actionable fixes.
  • Use Case: Assess a code file or dataset to uncover potential weaknesses and prioritize remediation.

Quick Start

Provide the target file to Pi-Kimi and obtain a JSON report with a score, a one-sentence summary, and identified issues.

Frequently Asked Questions about ai-pi-nv-kimi

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I perform an adversarial code review using Kimi K2.5?

To perform an adversarial code review, provide your target code or data to the Skill, which invokes Kimi K2.5 via the NVIDIA Pi integration to probe for vulnerabilities and output a structured JSON report for remediation.

What is an adversarial assessment and how does it identify code vulnerabilities?

An adversarial assessment probes targets to uncover weaknesses, generating a structured JSON object with a 0-1 score, a one-line summary, and identified issues to guide vulnerability remediation.

Can I use NVIDIA NIM to generate a structured vulnerability report for my dataset?

Yes, you can use this Skill with NVIDIA NIM to assess a dataset, orchestrating an adversarial evaluation that outputs a structured JSON score, summary, and issue list for actionable fixes.

How do I get a JSON score and summary for code weaknesses found in my files?

You get a JSON score and summary by providing your target file to the Skill, which parses the adversarial evaluation output and reports findings including a 0-1 score, one-sentence summary, and identified issues.

Does this adversarial review tool require any specific dependencies to run?

No specific dependencies are required to run this adversarial review tool, as it operates standalone to invoke Kimi K2.5 through the Pi integration and parse the resulting JSON findings.

What is the best way to prioritize remediation after an adversarial evaluation?

The best way to prioritize remediation is to analyze the structured JSON output, using the 0-1 score and identified issues list to quickly address the most critical vulnerabilities found during the adversarial assessment.