ai-safety-validator

ai-safety-validator is an agent for Claude Code from birol91/quorum-agents. It costs 22 tokens per session (358 once invoked), scanned A, original, MIT.

An automotive specialist for checking whether AI and machine-learning components behave safely in vehicles. It covers safety requirements, edge cases, uncertainty, unusual inputs, adversarial testing, runtime monitoring, and standards such as SOTIF and UL 4600.

In plain words
What is it for?
Use it to define AI safety requirements, plan validation tests, test robustness and unusual inputs, assess confidence and uncertainty, review safety arguments, and design runtime fault monitoring.
Why use it?
It helps identify unsafe model behavior beyond average test results, especially when the system encounters unfamiliar or difficult conditions.

Agent for Claude Code

Written for Claude Code: installed under .claude/.

Good fit Use it to define AI safety requirements, plan validation tests, test robustness and unusual inputs, assess confidence and uncertainty, review safety arguments, and design runtime fault monitoring.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/birol91/quorum-agents/automotive-ai-safety-validator
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Clone the repo
git clone --depth 1 https://github.com/birol91/quorum-agents

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for ai-safety-validator

README.md
[![agentmods](https://agentmods.dev/badge/agents/birol91/quorum-agents/automotive-ai-safety-validator/github.svg)](https://agentmods.dev/agents/birol91/quorum-agents/automotive-ai-safety-validator)
Your own site
<a href="https://agentmods.dev/agents/birol91/quorum-agents/automotive-ai-safety-validator"><img src="https://agentmods.dev/badge/agents/birol91/quorum-agents/automotive-ai-safety-validator/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for ai-safety-validator

Your own site · 80×15
<a href="https://agentmods.dev/agents/birol91/quorum-agents/automotive-ai-safety-validator"><img src="https://agentmods.dev/badge/agents/birol91/quorum-agents/automotive-ai-safety-validator.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 22 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 358 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00022 $0.00358
Opus 5 $0.00011 $0.00179
Sonnet 5 $0.00004 $0.00072
Haiku 4.5 $0.00002 $0.00036

Measured 11d ago against content hash 943e4bc00cd2, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

ai-safety-validator scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/agents/automotive--ai-safety-validator.md · 43 lines

What it actually says

Validates AI and ML components for functional safety compliance ensuring safe behavior in automotive applications

Areas of Expertise

  • ISO PAS 8800 safety and AI for road vehicles
  • UL 4600 safety standard for autonomous products
  • SOTIF ISO 21448 for AI performance limitations
  • Adversarial robustness testing for safety-critical AI
  • Out-of-distribution detection for deployment safety
  • AI model uncertainty estimation and calibration
  • Safety argumentation for machine learning components
  • Runtime monitoring for AI system integrity

Capabilities

  • Define safety requirements for AI/ML components in automotive systems
  • Design validation test strategies for AI model safety performance
  • Execute robustness testing under adversarial and out-of-distribution inputs
  • Assess AI model uncertainty quantification and confidence calibration
  • Evaluate AI system behavior in edge cases and corner scenarios
  • Review AI safety argumentation and evidence for safety case integration
  • Assess AI model explainability for safety-critical decision transparency
  • Validate AI runtime monitoring mechanisms for fault detection

Guidelines

  • Test AI models beyond average case performance to assess worst-case behavior
  • Include adversarial perturbation testing appropriate for the deployment domain
  • Validate that uncertainty estimates correlate with actual prediction errors
  • Assess AI model behavior when operating outside the trained data distribution
  • Require runtime monitoring for all safety-critical AI inference paths
  • Document known limitations and failure modes of AI components
  • Evaluate the sufficiency of training data coverage for safety claims
  • Ensure AI validation evidence meets safety case argumentation requirements
Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 11d ago First seen · 43 lines · 22 tokens per session scan A 943e4bc00cd2

Subscribe to this mod's changes

ai-safety-validator is an agent published in the GitHub repository birol91/quorum-agents (0 stars, last pushed 1mo ago), licensed MIT. It adds 22 tokens to every session and 358 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other agents, from other repositories