InnerVault

Library

85,762 resources from 11 public sources. 54 checked by hand; the rest are unreviewed.

Results

  1. SkillSpector

    Skills

    Security scanner for AI agent skills. Detect vulnerabilities, malicious patterns, security risks, prompt injection, data exfiltration, and supply-chain risks in Claude Code, Codex, and MCP skills before you install them.

    Indexed GitHub, NVIDIA 25.9k ★Apache-2.0Updated 2 days ago

    git clone https://github.com/NVIDIA/SkillSpector.git
  2. skillguard

    Skills

    Security scanner for AI agent skills. Detects prompt injection, data exfiltration, and malicious payloads before you install.

    Stale GitHub, obielin 20 ★Custom licence (see repository)Updated 24 days ago

    git clone https://github.com/obielin/skillguard.git
  3. little-canary

    Skills

    Detects prompt injection by its effect on a sacrificial canary model, not just pattern matching: untrusted input hits a powerless model first, a behavioral check reads the residue, and it returns block, flag, or pass before your primary model acts. Inbound preflight sensor, not a guarantee.

    Stale GitHub, hermes-labs-ai 39 ★Apache-2.0Updated 27 days ago

    git clone https://github.com/hermes-labs-ai/little-canary.git
  4. Sponsio

    Skills

    Deterministic safety solutions for probabilistic AI agents

    Stale GitHub, SponsioLabs 455 ★Apache-2.0Updated 21 days ago

    git clone https://github.com/SponsioLabs/Sponsio.git
  5. A curated corpus of incidents, attack vectors, failure modes, and defensive tools for autonomous AI agents.

    Stale GitHub, h5i-dev 57 ★MITUpdated 32 days ago

    git clone https://github.com/h5i-dev/awesome-ai-agent-incidents.git
  6. Browser automation with evidence. An Agent Skill that lets an AI agent drive a real browser while every action passes a policy gate, every run leaves an auditable artifact, and the run compiles into a plain Playwright script that replays for free.

    Indexed GitHub, OMD-123 13 ★MITUpdated 2 days ago

    git clone https://github.com/OMD-123/browser-verified-skill.git
  7. OpenART

    Skills

    OpenART is an open-source framework designed to evaluate the safety and robustness of autonomous AI agents in dynamic, long-horizon, and stateful environments. It stress-tests agent runtimes against multi-step state poisoning, privilege escalation, and tool-use vulnerabilities across 10,000+ benchmark scenarios.

    Stale GitHub, AI45Lab 263 ★AGPL-3.0Updated 15 days ago

    git clone https://github.com/AI45Lab/OpenART.git
  8. 5,660 AI security links: prompt injection, LLM & AI agent security, MCP, coding agents, agent skills & supply chain, red teaming, guardrails, sandboxing, incidents & CVEs, governance. 570 arXiv papers, 1,155 GitHub repos, 570 X posts, 29 categories of plain markdown you can grep, diff and fork. Obsidian vault export. Added, not endorsed.

    Stale GitHub, AnthonyHerman 8 ★CC-BY-4.0Updated 5 days ago

    git clone https://github.com/AnthonyHerman/ai-security-corpus.git
  9. A complete study toolkit for AI assistants — Claude, Codex, Gemini CLI, Cursor, Copilot: assignment coaching, professor-style grade estimates, academic writing, exam prep, lecture notes, study planning and Moodle/Canvas sync for every student.

    Stale GitHub, Noam-Marom 11 ★MITUpdated 5 days ago

    git clone https://github.com/Noam-Marom/academic-students.git
  10. Tinder for your Claude Code skills, subagents and MCP servers. Swipe away what wastes context, scan for prompt injection, get honest token costs. Fixes what it finds and actually deletes. Zero dependencies, Claude Code + Codex.

    Stale GitHub, khendzel 122 ★MITUpdated 47 days ago

    git clone https://github.com/khendzel/skills-janitor.git
  11. aizee

    Skills

    The policy layer for AI coding agents. aiZee governs what Cursor, Claude Code, Copilot, Windsurf, Devin & MCP agents may do — 139 enforcement gates, 29 personas, 245 skills, BMAD runtime, hybrid memory, tamper-evident audit. pip install aizee

    Indexed GitHub, m3taz-ahmed 6 ★MITUpdated yesterday

    git clone https://github.com/m3taz-ahmed/aizee.git
  12. Zero-LLM deterministic jailbreak regression benchmark: runs a repeatable battery of known-pattern attacks against an LLM endpoint and scores refusal/partial/compliance across runs. Catch when a model or prompt update silently got weaker. Single-turn, responsible-use. pip install hermes-jailbench

    Stale GitHub, hermes-labs-ai 12 ★Apache-2.0Updated 27 days ago

    git clone https://github.com/hermes-labs-ai/hermes-jailbench.git
  13. agent-guard

    Skills

    Scan AI agent skills, plugins, MCP servers, and CLI tools (npm/PyPI/Go, cargo, release binaries, curl|bash installers) for malicious code before installation - one scan, every agent (Claude Code, Codex, Gemini, Hermes, OpenClaw). Powered by NVIDIA SkillSpector, Cisco mcp-scanner, Datadog GuardDog, VirusTotal + malcontent.

    Indexed GitHub, elliottwaves-20 5 ★MITUpdated yesterday

    git clone https://github.com/elliottwaves-20/agent-guard.git
  14. bounty-guard

    Skills

    Zero-dependency Python CLI, GitHub Action, and MCP server scanning untrusted GitHub issues and repositories for prompt injection, CI secret exfiltration, and cryptographic bounty escrow.

    Stale GitHub, doomsday-network 15 ★MITUpdated 22 days ago

    git clone https://github.com/doomsday-network/bounty-guard.git

Indexed entries aren't reviewed or tested by InnerVault. Details come straight from each source. Anything that can reach your accounts, files or browser can do damage, so read the source before you install it.