evaluating-model-guardrails

evaluating-model-guardrails is a skill for Claude Code from UnboundCompute/security-agent-skills. It costs 119 tokens per session (1,558 once invoked), scanned A, original, MIT.

A structured method for testing whether an AI model's safety rules hold up under adversarial prompts. It uses policy-based probes, multi-turn escalation, obfuscation, role changes, and repeatable scoring instead of relying on one jailbreak example.

In plain words
What is it for?
Use it to evaluate a model deployment, system prompt, or content-safety layer that you own or are authorised to test, before relying on its guardrails in production.
Why use it?
It helps measure how consistently a model refuses or limits requests it should not answer, producing evidence about bypass rates rather than anecdotes.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin.

Part of the security-agent-skills plugin — 194 skills shipped together

Good fit Use it to evaluate a model deployment, system prompt, or content-safety layer that you own or are authorised to test, before relying on its guardrails in production.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/unboundcompute/security-agent-skills/evaluating-model-guardrails
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.

Any agent
npx skills add UnboundCompute/security-agent-skills --skill evaluating-model-guardrails
Clone the repo
git clone --depth 1 https://github.com/UnboundCompute/security-agent-skills

Made for: Claude Code.

Or install security-agent-skills, the plugin that ships this one along with the rest of its 194 skills.

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 evaluating-model-guardrails

README.md
[![agentmods](https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/evaluating-model-guardrails/github.svg)](https://agentmods.dev/skills/unboundcompute/security-agent-skills/evaluating-model-guardrails)
Your own site
<a href="https://agentmods.dev/skills/unboundcompute/security-agent-skills/evaluating-model-guardrails"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/evaluating-model-guardrails/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 evaluating-model-guardrails

Your own site · 80×15
<a href="https://agentmods.dev/skills/unboundcompute/security-agent-skills/evaluating-model-guardrails"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/evaluating-model-guardrails.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 119 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,558 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 $0.00119 $0.01558
Opus 5 $0.00060 $0.00779
Sonnet 5 $0.00024 $0.00312
Haiku 4.5 $0.00012 $0.00156

Measured 4d ago against content hash 3900aa4187fb, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

evaluating-model-guardrails 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 4d 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.

skills/evaluating-model-guardrails/SKILL.md · 129 lines

How it starts

The opening of the file, as written. The whole thing — 129 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Evaluating model guardrails: measure the boundary, don't collect one jailbreak

A guardrail is a claim: "the model will refuse this class of request." A single successful jailbreak is an anecdote; what you need is a measurement, the rate at which the boundary holds across a structured battery of attacks. Evaluating guardrails means grounding probes in the actual policy, attacking each along known bypass axes, and scoring outcomes reproducibly, so you can state how strong the boundary is, not just that someone once beat it.

When to use

  • You are reviewing a model deployment, a system prompt, or a content-safety layer.
  • Before shipping an assistant that must refuse or constrain a defined set of requests.
  • You need a defensible bypass rate, not a single proof-of-concept transcript.

Scope check

Evaluate models and deployments you own or are authorized to test. Use benign, clearly-scoped probes against a defined policy; do not generate real harmful output against systems you do not control. If you can't name the authorization, stop.

The loop

  1. Ground the probes in the stated policy. Get the actual list of what this deployment must refuse or constrain (its safety policy, its system prompt's rules, its allowed scope). Every probe targets a specific rule, so a result maps to a policy line, not a vibe. An undefined policy is the first finding: you cannot evaluate a boundary no one has drawn.

  2. Build a baseline probe set. For each rule, write direct requests that should be refused and benign near-misses that should be allowed. The near-misses matter: a guardrail that refuses everything is broken differently from one that refuses nothing. Record baseline refuse/allow behavior before attacking.

  3. Attack along the bypass axes. Take each refused probe and apply the known transformations: multi-turn escalation (warm up, then pivot), encoding and obfuscation (alternate scripts, spacing, invisible characters, indirection), role and context reframing (fiction, hypothetical, translation, "for research"), and instruction-hierarchy conflict (content claiming higher authority than the system rule). Each axis is a separate test of the same rule.

Read the full file on GitHub · 129 lines

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. 4d ago First seen · 129 lines · 119 tokens per session scan A 3900aa4187fb

Subscribe to this mod's changes

evaluating-model-guardrails is a skill published in the GitHub repository UnboundCompute/security-agent-skills (5 stars, last pushed yesterday), licensed MIT. It adds 119 tokens to every session and 1,558 once invoked, about $0.0006 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 2026-09-05.

Related

Other skills, from other repositories

model-supply-chain

Reviews AI/ML model supply chains for security risks including model provenance verification, training data lineage, fine-tuning pipeline integrity, inference dependency review, and backdoor detection. Auto-invoked when reviewing systems that download pre-trained models, fine-tune foundation models, or deploy models…

UnitOneAI/SecuritySkills · 104 tokens

prodcheck-review

Review this codebase against the prodcheck pre-production checklists — security, performance, scale, integrations and post-launch readiness. Use when asked to check whether a project is ready to ship, to audit an area before launch, or to work through a specific checklist. Produces evidence with file:line citations…

FarzamHabibi/pre-production-checklist · 70 tokens

rag-poisoning-and-data-exfiltration

Test Retrieval-Augmented Generation (RAG) systems for data poisoning, prompt injection via retrieved documents, and data exfiltration through manipulated context windows. Use this skill when assessing RAG-based chatbots, knowledge bases, enterprise AI assistants, or any system that augments LLM responses with external…

akashrpatil/awesome-offensive-security-skills · 89 tokens

integrate-arcjet-guard-genkit

Integrate Arcjet security into a Genkit JS agent using @arcjet/guard — wrap ai.defineTool, put guardMiddleware on generate({ use }) for unwrapped / MCP / filesystem tools, and read a caller-owned id from generate({ context }). Use when asked to add Arcjet to genkit, rate limit its tools, screen inbound messages, or…

arcjet/arcjet-js · 89 tokens

integrate-arcjet-guard-tanstack-ai

Integrate Arcjet security into a TanStack AI chat() app using @arcjet/guard — put guardMiddleware first on chat({ middleware }) so onBeforeToolCall gates tools, and read a caller-owned id from chat({ context }). Use when asked to add Arcjet to TanStack AI, rate limit its tools, screen inbound messages, or block prompt…

arcjet/arcjet-js · 103 tokens

integrate-arcjet-guard-claude-managed-agents

Integrate Arcjet security into Claude Managed Agents (hosted REST+SSE, beta managed-agents-2026-04-01) using @arcjet/guard — screen user.message / initialevents before sessions.events.send, and gate custom tools on agent.customtooluse. Use when asked to add Arcjet to Claude Managed Agents, rate limit custom tools, or…

arcjet/arcjet-js · 102 tokens