convez.ai
MCP ServerSegurançaAvançado
LLMSecurity

Awesome Agent Skills Security

LLMSecurity/awesome-agent-skills-security

Referência completa sobre segurança de skills e ferramentas em agentes IA

Ver no GitHub
BomPluginScore · de 0 a 100

Uma curadoria abrangente de recursos sobre segurança de habilidades em agentes IA, cobrindo ataques, defesas, frameworks e benchmarks. Estrutura conhecimento sobre a superfície de ataque criada pelo uso de ferramentas externas, plugins e skills em sistemas agentic.

Instalação

$npx convez-ai LLMSecurity/awesome-agent-skills-security

1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.

Métricas

Estrelas
288
Forks
142
Acompanhando
5
Saúde
62%
Último commit
há 1 mês
Versão
—
Issues abertas
38
Contribuidores
11

Casos de uso

  • Pesquisa e desenvolvimento em segurança de agentes IA
  • Implementação de controles de segurança em sistemas agentic
  • Avaliação de riscos em arquiteturas com tool use
  • Red teaming e teste de robustez de agentes

Veredito

Recurso essencial e bem estruturado para arquitetos e pesquisadores em IA focados em segurança agentic. Mapeia sistematicamente a paisagem de ameaças emergentes e defesas em um domínio crítico e ainda pouco documentado na indústria.

README

A tradução para o português deste README ainda está na fila.

Awesome Agent Skills Security Awesome

🛡️ A curated list of resources on securing AI agent tool use and skill ecosystems — attacks, defenses, frameworks, benchmarks, and standards.

AI agents increasingly use external tools, plugins, and skills to interact with the world. This creates a new attack surface: agent skills security. This list covers the threats, defenses, and research landscape for securing these capabilities.

Contents


Threat Frameworks & Standards

Surveys & Systematizations

Attack Research

Prompt Injection via Tools

Tool Poisoning & Supply Chain

Privilege Escalation & Excessive Agency

Data Exfiltration & Privacy

Indirect Prompt Injection

Agent Deception & Manipulation

Compound System Attacks

Cross-Plugin Attacks

Backdoor Attacks on Agents

Jailbreaking & Guardrail Bypass

Defense Research

Permission & Access Control

Runtime Monitoring & Sandboxing

Input/Output Validation

Formal Verification & Analysis

Evaluation & Red Teaming

Benchmarks & Datasets

BenchmarkFocusSizePaper
ASBComprehensive agent security10 agents, 398 envsZhang et al.
InjecAgentIndirect prompt injection1,054 test casesZhan et al.
R-JudgeSafety risk awareness162 records, 27 scenariosYuan et al.
ToolSwordTool learning safety6 scenarios, 3 stagesYe et al.
AgentDynDynamic prompt injectionOpen-ended, extensibleLi et al.
SkillSafetyBenchSkill-mediated agent safety155 cases, 47 tasksJin et al.
TABSelective cue following in terminal agents89 terminal tasksMavali et al.
Skill-InjectSkill file attacksMulti-scenarioSchmotz et al.
NAAMSEEvolutionary agent security evalAdaptive red-teamingPai et al.
AgentHarmAgent misuse110 behaviors, 440 variantsAndriushchenko et al.
SkillGuard DatasetMalicious skill detection157 malicious skillsLiu et al.
WIPIWeb-based indirect injectionMulti-scenarioLiu et al.

Tools & Frameworks

ToolDescriptionLink
Bounty SieveOffline-by-default bounty intake guardrail for coding agents: read-only GitHub issue/URL-list import, deterministic triage, local decision briefs, and human approval gatesGitHub
SkillGuardLLM-native agent skill security auditor (OWASP Agentic + MITRE ATLAS)GitHub
PipelockOpen-source AI agent firewall and MCP-aware egress proxy with DLP, prompt injection scanning, process sandboxing, and mediator-signed action receiptsGitHub
NemoClawNVIDIA reference stack for running always-on AI agents more safely in sandboxes, with network policy, hardening, routed inference, and lifecycle controlsGitHub
Invariant GuardrailsPolicy-based agent security guardrailsGitHub
Armorer GuardLocal Rust scanner for AI-agent prompt injection, credential redaction, sensitive-data requests, exfiltration-style text, and dangerous tool-call contextGitHub
LLM GuardInput/output scanning for LLM applicationsGitHub
RebuffSelf-hardening prompt injection detectorGitHub
NeMo GuardrailsNVIDIA's toolkit for adding guardrails to LLM-based applicationsGitHub
Lakera GuardEnterprise prompt injection defense APIWebsite
PromptfooLLM red teaming and evaluation frameworkGitHub
GarakLLM vulnerability scannerGitHub
IPI-ProxyIntercepting proxy for red-teaming web-browsing agents against indirect prompt injection on live whitelisted domainsGitHub
Tuning Engines CLIMCP server and CLI for governed agent/skill/tool access with policy checks, approvals, traces, and role-scoped registriesGitHub
AgentSkillsScannerStatic analysis scanner for agent skill definitionsGitHub
SkilLockBehavior-pinning lockfile + capability-delta PR review for Claude Code & Codex skills; SARIF output for Code ScanningGitHub
agent-diff-guardPre-push guardrail that flags high-risk coding-agent diffs such as CI/CD changes, dependency edits, test deletions, hardcoded secrets, and task-scope drift before mergeGitHub
SkillidPolicy-driven Claude Code plugin that combines skill guidance with per-tool hooks to enforce org guardrails, confirmation rules, redaction, and connector-specific access controlGitHub
Agent AuditSecurity analysis system for LLM agent apps: dataflow analysis, credential detection, MCP config parsing, privilege-risk checksZhang et al.
mcp-sec-auditMCP server security toolkit: static pattern matching + dynamic sandboxed fuzzing via Docker/eBPF for detecting over-privileged tool capabilitiesHuang et al.
Assay HarnessCI gate that checks an agent's claimed tool side-effects (filesystem, network, process) against independently observed runtime evidence, classifying each claim as supported, degraded, blocked, or not-evaluable (observed support is the ceiling)GitHub
Agent ScanSnyk's scanner for local agent supply chains, covering MCP servers and skills with checks for prompt injection, tool poisoning, toxic flows, and malware-laced skill filesGitHub
DScanOpen-source agent security suite for runtime tool-call tracing, prompt-injection shielding, MCP audits, adversarial testing, and sequence-level attack detectionGitHub
Sayfos SDKRuntime guardrail SDK for AI agents with provenance checks, budget governance, plan preflight, and adjudication tokens before high-risk tool actionsGitHub
agent-sentineleBPF/BPF-LSM prototype that monitors local agent behavior and atomically blocks prompt-injection-driven access to sensitive files at the kernel boundaryGitHub
mcp-sploitMetasploit-style framework for authorized security testing of MCP servers and MCP security gateways, including enumeration and exploit modules for unsafe tool exposureGitHub
ClawvisorAI agent gateway for purpose-based authorization, credential vaulting, and audit logging — agents declare task scope, humans approve once, Clawvisor enforces on every request without the agent ever seeing credentialsGitHub
trentclawSecurity assessment skill for OpenClaw environments: scans gateway config, skill permissions, MCP trust boundaries, and plugins, and correlates them into chained attack paths with severity-ranked remediation stepsGitHub

Agent Skill Specifications

SpecificationOrgFocus
AgentSkills.ioOpen StandardAgent skill definition and security requirements
Model Context Protocol (MCP)AnthropicTool/resource integration protocol for LLMs
OpenAI Function CallingOpenAITool use specification for GPT models
Tool Use (Claude)AnthropicClaude's native tool use interface
LangChain ToolsLangChainTool abstraction for agent frameworks
AutoGPT PluginsAutoGPTPlugin system for autonomous agents
OpenAPI/SwaggerLinux FoundationAPI specification commonly used as tool definitions

Industry Reports & Blog Posts

Related Awesome Lists

Contributing

Contributions are welcome! Please read the contribution guidelines before submitting a pull request.

How to Contribute

  1. Fork the repository
  2. Add your resource in the appropriate category
  3. Use the format: - 📄 **[Title](URL)** — Authors, Venue Year. One-sentence description.
  4. Submit a pull request

Criteria

  • Resources must be directly related to agent/tool/skill security
  • Papers should be published or on arXiv
  • Tools should be actively maintained (commits within last 6 months)
  • Blog posts should provide substantial technical analysis

Citation

If you find this list useful in your research, please cite:

@misc{awesome-agent-skills-security,
  author = {Liu, Yi},
  title = {Awesome Agent Skills Security},
  year = {2026},
  publisher = {GitHub},
  journal = {GitHub Repository},
  howpublished = {\url{https://github.com/LLMSecurity/awesome-agent-skills-security}}
}

License

CC0

This list is released under CC0 1.0 Universal.

Ver o original no GitHub

Bom para

Relacionados