safety-evaluation

safety-evaluation is a skill for Claude Code from jassics/awesome-claude-security. It costs 71 tokens per session (653 once invoked), scanned A, original, GPL-3.0.

A method for testing whether an AI model or feature behaves safely across different types of harmful requests and factual answers.

In plain words
What is it for?
Use it to measure refusals, resistance to problematic inputs, unnecessary refusals, helpfulness, and whether answers stay grounded in the available information.
Why use it?
It provides repeatable tests, scoring rules, and pass-or-fail limits instead of relying on informal checks before release.

Skill for Claude Code

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

Part of the ai-safety plugin — 6 skills, 1 agent shipped together

Good fit Use it to measure refusals, resistance to problematic inputs, unnecessary refusals, helpfulness, and whether answers stay grounded in the available information.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/jassics/awesome-claude-security/safety-evaluation
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 jassics/awesome-claude-security --skill safety-evaluation
Clone the repo
git clone --depth 1 https://github.com/jassics/awesome-claude-security

Made for: Claude Code.

Or install ai-safety, the plugin that ships this one along with the rest of its 6 skills, 1 agent.

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 safety-evaluation

README.md
[![agentmods](https://agentmods.dev/badge/skills/jassics/awesome-claude-security/safety-evaluation/github.svg)](https://agentmods.dev/skills/jassics/awesome-claude-security/safety-evaluation)
Your own site
<a href="https://agentmods.dev/skills/jassics/awesome-claude-security/safety-evaluation"><img src="https://agentmods.dev/badge/skills/jassics/awesome-claude-security/safety-evaluation/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 safety-evaluation

Your own site · 80×15
<a href="https://agentmods.dev/skills/jassics/awesome-claude-security/safety-evaluation"><img src="https://agentmods.dev/badge/skills/jassics/awesome-claude-security/safety-evaluation.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 71 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 653 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 unknown 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 $0.00071 $0.00653
Opus 5 $0.00036 $0.00327
Sonnet 5 $0.00014 $0.00131
Haiku 4.5 $0.00007 $0.00065

Measured 9d ago against content hash f2df721a0e10, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

safety-evaluation 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 9d 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.

plugins/ai-safety/skills/safety-evaluation/SKILL.md · 57 lines

The source is not reproduced here

Licensed GPL-3.0

The repository is licensed GPL-3.0, which this catalogue does not treat as permission to reproduce the file. Read it at the source.

Read it on GitHub

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. 9d ago First seen · 57 lines · 71 tokens per session scan A f2df721a0e10

Subscribe to this mod's changes

safety-evaluation is a skill published in the GitHub repository jassics/awesome-claude-security (6 stars, last pushed 1mo ago), licensed GPL-3.0. It adds 71 tokens to every session and 653 once invoked, about $0.0004 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 skills, from other repositories

goal-test

A local experiment for testing a goal command that keeps an AI coding session working until a stated condition is judged complete. It uses a separate language model to evaluate the conversation after each assistant turn.

restarter/lets-workflow · 137 tokens

prompt-testing

Use when comparing two prompt variants, defining quality/efficiency/robustness metrics, or deciding whether to adopt a challenger prompt over a baseline.

fusengine/agents · 33 tokens

mutation-test

Mutation testing with two engines. Uses the project's NATIVE mutation runner (StrykerJS / Infection / mutmut / PIT / cargo-mutants) when one is configured — installing it on explicit consent when it is not — for a reproducible, comparable score; and an LLM-guided engine for the mutation classes native mutators cannot…

greglas75/zuvo · 190 tokens

deep-plan

Creates detailed, sectionized, TDD-oriented implementation plans through research, stakeholder interviews, and multi-LLM review. Use when planning features that need thorough pre-implementation analysis.

piercelamb/deep-plan · 39 tokens

eval-set

Build and run a project-specific retrieval eval so changes to your rules or base prompt are scored, not eyeballed. Mirrors tests/groundtruth.json + clawness eval: you write prompt→expected-rule cases, then measure MRR@k and hit-rate before and after an edit. Run it after trimming a base prompt into ranked retrieval…

fullymiddleaged/Clawness · 98 tokens

data-pipeline-quality

Automated data quality checks for pipelines. Testing pyramids, dbt test patterns, data contracts, circuit breakers, and monitoring. Use when implementing data quality checks, writing dbt tests, defining data contracts, setting up pipeline validation, building automated quality monitoring, or when someone asks "how do…

hollandkevint/data-product-operator · 88 tokens