Best prompt injection and agent security skills, page 3

Audit raytsystem changes for prompt injection, provenance bypass, path/symlink/hardlink escape, secret leakage, stale fencing, partial promotion, unsafe parsing, and unapproved side effects.

romarayt/raytsystem-public-os

Extract structured data via stored browser-templates or one-shot DOM queries, with mandatory AIDefence PII + prompt-injection gates before content reaches the model

ruvnet/ruflo

Threat-model and harden AI agents, RAG systems, assistants, and tool-using workflows against direct, indirect, stored, cross-agent, and multimodal prompt injection.

seb1n/awesome-ai-agent-skills

Verify exact SillyTavern, Tavern Helper / JS-Slash-Runner, STScript, macro, prompt-injection, worldbook, EJS, MVU, and runtime-library capabilities before implementing or reviewing rolecard…

LiarMTTT/TavernWeave

Use before committing changes to auth, workspace scoping, channel webhooks, AI tools/MCP, permission settings, or anything handling untrusted channel content in ChatbotX.

ChatbotXIO/ChatbotX

Defend AI systems against prompt injection and indirect prompt attacks using input controls, tool permissions, output validation, and isolation boundaries.

sickn33/agentic-awesome-skills

Scan inputs for prompt injection, unsafe content, and adversarial attacks using AIDefence.

ruvnet/ruflo

Screen fetched web pages, tool results, emails and file contents for instructions aimed at the agent before they enter context, using a keyless multi-label classifier over chunks.

mrmps/classifier-dev

[omh] Agent or automation safety risks: review prompt, tool, secret, dependency, destructive-action, and explicit local plugin risks before agent or code execution.

rlaope/oh-my-hermes

AI/LLM defensive security reference: prompt-injection defense, OWASP LLM Top 10 defensive mapping, MCP and agentic tool-call hardening, training-data poisoning detection, model-output validation and…

modu-ai/moai-adk

DevSecOps, container, and API operational defensive security reference: CI/CD pipeline hardening, secret scanning, IaC misconfiguration detection, SAST/DAST integration, container image scanning…

modu-ai/moai-adk

Run and inspect the local pieces of Guardana — the throwaway PostgreSQL for the collector, the collector itself, a fake OpenAI-compatible endpoint to probe, the documentation site served locally…

guardana/guardana

Authorized security assessment of LLM applications and AI agents: prompt injection, tool abuse, RAG exposure, memory poisoning, system-prompt extraction, and agent-compliance engineering per OWASP…

sickn33/agentic-awesome-skills

Detects prompt injection using regex signature matching, heuristic scoring for structural anomalies, and DeBERTa-based transformer classification, flagging direct injections (system-prompt…

mukul975/Anthropic-Cybersecurity-Skills

Detect and defend against indirect prompt injection hidden in web pages, documents, and images consumed by an agent, via content extraction (HTML/PDF/OCR), normalization, and scanning with LLM…

mukul975/Anthropic-Cybersecurity-Skills

Runs NVIDIA garak probe suites (jailbreak, prompt injection, data leakage, toxicity, and more) against an LLM endpoint - Hugging Face models, OpenAI-compatible APIs, or Bedrock - then interprets the…

mukul975/Anthropic-Cybersecurity-Skills

Deploys Llama Guard 3 safety classification, NeMo Guardrails programmable dialogue rails, and LLM Guard input/output scanner pipelines as complementary runtime defenses that inspect and constrain…

mukul975/Anthropic-Cybersecurity-Skills

Runtime security guardian for OpenClaw agents. An agent skill from LeoYeAI/openclaw-master-skills.

LeoYeAI/openclaw-master-skills

This skill should be used when the user asks to "scan AI systems for security threats", "check for prompt injection vulnerabilities", "assess model security posture", "detect data poisoning risks"…

borghei/Claude-Skills

A skill your agent uses when assessing AI/ML systems for prompt injection, jailbreak vulnerabilities, model inversion risk, data poisoning exposure, or agent tool abuse.

alirezarezvani/claude-skills

Consolidate brain memory and mine user sessions since the last consolidation checkpoint (sleep-cycle style).

mikeyobrien/rho

Binance Web3 official skill — security audit for token contracts, detecting honeypots, rug pulls, and malicious functions across BSC, Base, Solana, and Ethereum.

TermiX-official/cryptoclaw

Pull AWS Security Agent findings (penetration tests and code reviews) and drive remediation.

aws/agent-toolkit-for-aws

Behavioral trust grades (A–F) for MCP servers. An agent skill from BankrBot/skills.

BankrBot/skills

AI Agent 安全开发与防护最佳实践，包含prompt注入防护、代码执行安全、敏感信息保护、合规审计全流程规范. An agent skill from ProgrammerAnthony/Expert-Coding-Harness.

ProgrammerAnthony/Expert-Coding-Harness

LLM / AI application attack hunting - prompt injection (direct + indirect), excessive agency, insecure output handling, system-prompt + data leakage.

Encod3d-Sec/TORCH

MCP server attack hunting - tool poisoning, indirect prompt injection via tool output, rug-pull updates, cross-tool shadowing, over-permissioned/excessive-agency tools, lethal trifecta.

Encod3d-Sec/TORCH

Test LLM-integrated applications against known prompt injection techniques, evasion methods, and attack intents using the Arcanum PI Taxonomy.

OWASP/secure-agent-playbook

Audit an LLM application for indirect prompt injection - compose objective x technique payloads, deliver them through the channels the agent actually reads, and prove impact with an out-of-band…

forefy/.context

Defensive security engineering judgment, distilled from a stronger model - invoke when THREAT MODELING a system or feature; making security-relevant design decisions (auth, crypto, trust boundaries…

telagod/code-abyss

Audit MCP servers for tool poisoning, tool shadowing, rug pulls, SSRF, and unauthenticated exposure using Invariant Labs' mcp-scan for static/runtime scanning plus manual SSRF/auth checks and…

mukul975/Anthropic-Cybersecurity-Skills

Test whether authorized direct or indirect untrusted content can alter an AI system's protected behavior, context use, memory, retrieval, output handling, or downstream capability.

cyberful/cyberful

Reconstruct how instructions, retrieved content, memory, identities, approvals, tool schemas, arguments, outputs, delegation, and fallbacks propagate through an AI system.

cyberful/cyberful

Systematic diagnosis of a red test, a rule that fires or stays silent wrongly, a scan or probe whose artifact looks wrong, a collector error, a red CI run or a gate that is green for the wrong reason.

guardana/guardana

Implement content safety guardrails for Claude — input filtering, Use when working with policy-guardrails patterns.

jeremylongshore/tons-of-skills-marketplace

Secure your Anthropic integration — API key management, input validation, Use when working with security-basics patterns.

jeremylongshore/tons-of-skills-marketplace

Security and compliance review framework for Kling AI integrations.

jeremylongshore/tons-of-skills-marketplace

Audit a Claude/Agent SKILL.md (or any AI skill / system prompt) for safety before installing or merging it.

mohitagw15856/pm-claude-skills

AI/LLM 间接 Prompt 注入攻击。当目标 AI 系统会处理外部数据源（网页、文档、邮件、数据库、API 返回）时使用。覆盖间接注入、工具链劫持、RAG 投毒、数据外泄等技术。OWASP LLM Top 10 1 漏洞类别

wgpsec/AboutSecurity

Adversarial defense layer for the mortgage plugin — protects against prompt injection, system prompt extraction, PII leakage, workflow bypass, and social engineering attacks.

davepoon/buildwithclaude

服务上线前的 AI 安全审查，聚焦需要理解代码语义和业务逻辑的安全问题. An agent skill from 312362115/claude.

312362115/claude

A skill your agent uses when auditing an ICML main-track submission for OpenReview, LaTeX formatting, 8-page body, anonymity, supplementary material, impact statement, dual submission, concurrent…

franklee16/academic-research-skills

Detects prompt injection hidden in documents from the other side (pleadings, skeletons, bundles, served evidence, opponents' emails and attachments) before an AI reads them, so the model is not…

lawve-ai/awesome-legal-skills

Route broad AI-agent security reviews to focused Cyberful skills across risk, model supply chain, context and capabilities, prompt injection, tool authorization, and RAG isolation.

cyberful/cyberful

Guide an OpenART agent or contributor through planning, running, extending, and debugging the framework.

AI45Lab/OpenART

Secure key management for AI agents. An agent skill from LeoYeAI/openclaw-master-skills.

LeoYeAI/openclaw-master-skills

Build AI agents with console.agent() - the jQuery of AI Agents.

LeoYeAI/openclaw-master-skills

A skill your agent uses when enforcing spend limits on AI agent wallets, validating transactions before signing, configuring allowlists or approval workflows, detecting prompt injection in agent…

LeoYeAI/openclaw-master-skills