llm-security

llm-security is a skill for Claude Code, Codex from Strategic-Automation/violin. It costs 16 tokens per session (426 once invoked), scanned A, original, MIT.

A security-testing guide for AI applications, including chatbots and agents, that checks prompt manipulation and Model Context Protocol (MCP) tool connections. MCP is a way for an AI agent to use tools and read resources.

In plain words
What is it for?
Testing prompt injection, jailbreaks, indirect attacks, and MCP or JSON-RPC tools, resources, and sessions with read-only checks.
Why use it?
It helps find ways users or outside content could trick an AI into ignoring its instructions or using tools improperly. It keeps this work separate from testing ordinary web APIs.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Testing prompt injection, jailbreaks, indirect attacks, and MCP or JSON-RPC tools, resources, and sessions with read-only checks.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/strategic-automation/violin/llm-security
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 Strategic-Automation/violin --skill llm-security
Clone the repo
git clone --depth 1 https://github.com/Strategic-Automation/violin

Made for: Claude Code, Codex.

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 llm-security

README.md
[![agentmods](https://agentmods.dev/badge/skills/strategic-automation/violin/llm-security/github.svg)](https://agentmods.dev/skills/strategic-automation/violin/llm-security)
Your own site
<a href="https://agentmods.dev/skills/strategic-automation/violin/llm-security"><img src="https://agentmods.dev/badge/skills/strategic-automation/violin/llm-security/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 llm-security

Your own site · 80×15
<a href="https://agentmods.dev/skills/strategic-automation/violin/llm-security"><img src="https://agentmods.dev/badge/skills/strategic-automation/violin/llm-security.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 16 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 426 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. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
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 $0.00016 $0.00426
Opus 5 $0.00008 $0.00213
Sonnet 5 $0.00003 $0.00085
Haiku 4.5 $0.00002 $0.00043

Measured 10d ago against content hash 355a9a93646b, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

llm-security 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 10d 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/llm-security/SKILL.md · 42 lines

What it actually says

LLM and Agent Security Testing

On-demand skill for AI/LLM application attack surfaces: prompt injection, jailbreaking, and Model Context Protocol (MCP) / JSON-RPC tool and resource exposure. It is distinct from REST/GraphQL API testing; MCP is an agent tool protocol, not a conventional API.

When to Use

  • A chatbot, recommendation, generation, or LLM-backed endpoint is in scope.
  • A Model Context Protocol endpoint (initialize, tools/list, tools/call, resources/read) or JSON-RPC surface is discovered.
  • Do not use for REST/GraphQL/SOAP inventory or authz; use api-testing for those, then route the confirmed class back to the right specialist playbook.

Playbook Routing

Class Playbook
LLM prompt injection, jailbreak, indirect injection playbooks/llm-prompt-injection.md
MCP / JSON-RPC tool, resource, and session testing playbooks/mcp-api-testing.md

Common Pitfalls

  • Identify the LLM integration and the tool/resource surface before injecting.
  • Keep probes read-only and reversible; do not extract or exfiltrate production secrets beyond what proves the flaw.
  • Treat a tool that is callable without its expected authorization as a broken-access-control candidate, not merely a protocol finding.

Verification Checklist

  • Engagement bootstrap and scope validation succeeded.
  • The matched playbook alone was loaded and bound to the PTT task.
  • Evidence captures the injected payload, the decisive response, and the intended versus observed behavior.
Files

What ships with it

2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 10d ago First seen · 42 lines · 16 tokens per session scan A 355a9a93646b

Subscribe to this mod's changes

llm-security is a skill published in the GitHub repository Strategic-Automation/violin (87 stars, last pushed 6d ago), licensed MIT. It adds 16 tokens to every session and 426 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 2026-08-30.

Related

Other skills, from other repositories

performing-jwt-none-algorithm-attack

Execute and test the JWT none algorithm attack to bypass signature verification by manipulating the alg header field in JSON Web Tokens.

xalgorix/xalgorix · 34 tokens

Web Application Security Testing

OWASP Top 10 testing, injection vulnerability detection, API security assessment, authentication testing, and web vulnerability reporting for authorized assessments.

Masriyan/Claude-Code-CyberSecurity-Skill · 30 tokens

bypassing-authentication-with-forced-browsing

Discovering and accessing unprotected pages, APIs, and administrative interfaces by enumerating URLs and bypassing authentication controls during authorized security assessments.

26zl/cybersec-toolkit · 39 tokens

kali-pentest-zh

A Chinese-language guide for carrying out authorized penetration tests with Kali Linux command-line tools through a local system, SSH, or Docker.

x-glacier/kali-pentest · 87 tokens

kali-pentest

Execute authorized penetration testing via Kali Linux CLI tools over SSH or Docker. Covers: information gathering, vulnerability analysis, sniffing & spoofing, web/API testing, exploitation, password attacks, wireless, cloud-native security, RFID/NFC, VoIP/ICS, reverse engineering, forensics, post-exploitation/C2, and…

x-glacier/kali-pentest · 73 tokens

api-security-testing

Security-test a REST, GraphQL, or gRPC API with Strix — autonomous agents that enumerate endpoints from an OpenAPI/GraphQL schema (or by crawling), then actually exploit the API-specific vulnerability classes in the OWASP API Security Top 10 (2023) — broken object-level authorization (BOLA/IDOR), broken object…

usestrix/strix · 144 tokens