llm-vendor-evidence-review

Review foundation model vendor evidence packs against firm deployment criteria.

Updated May 9, 2026
One-click install
npx skills add https://github.com/anotb/second-line-financial-services --skill llm-vendor-evidence-review
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: llm-vendor-evidence-review
Source: https://github.com/anotb/second-line-financial-services/tree/main/plugins/capability-plugins/ai-governance-model-risk/skills/llm-vendor-evidence-review
Command: npx skills add https://github.com/anotb/second-line-financial-services --skill llm-vendor-evidence-review

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill automates the review of foundation model vendor evidence packs, ensuring compliance with firm criteria and mitigating risks associated with model deployment.

Core Features & Use Cases

  • Automated Evidence Review: Reviews vendor evidence packs against firm criteria for deployment context.
  • Sufficiency View: Provides a sufficiency view by criterion, named gaps with supplemental evidence requested, and residual reliance with caveats.
  • Recommended Actions: Generates recommended owner actions and re-review triggers.
  • Use Case: Imagine you are deploying a foundation model for claims summarisation. Use this Skill to review the vendor's evidence pack against your firm's criteria, ensuring compliance and risk mitigation.

Quick Start

Run the llm-vendor-evidence-review skill with the following command: llm-vendor-evidence-review --model <model_name> --version <model_version> --deployment-form <deployment_form>.

Frequently Asked Questions about llm-vendor-evidence-review

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

FAQPage Schema
How do I automate vendor evidence review for foundation model deployment?

You can automate vendor evidence review by running the skill with the model name, version, and deployment form. It analyzes the evidence pack against firm criteria, generating a sufficiency view and identifying gaps for supplemental evidence requests.

What is model risk mitigation in vendor diligence workflows?

Model risk mitigation in vendor diligence involves analyzing regulatory material and applying conditional sector overlays to assess evidence sufficiency. It ensures operational resilience by structuring residual reliance with caveats before model deployment.

Can I use this for vendor diligence in regulated sectors?

Yes, vendor diligence for regulated sectors is supported through conditional sector overlays. The review applies regulatory and standards material analysis to ensure evidence sufficiency and compliance for your specific deployment context.

How do I generate recommended actions after reviewing an evidence pack?

Recommended owner actions are generated automatically after evaluating vendor evidence against firm criteria. The output includes specific actions and re-review triggers to address named compliance gaps and residual reliance caveats.

What does a sufficiency view by criterion show for vendor compliance?

A sufficiency view by criterion shows whether vendor evidence meets firm requirements for deployment. It highlights named gaps, requests supplemental evidence, and documents residual reliance with operational caveats for risk mitigation.