SEC·01★ 895
AgentShield
Adds guardrails and security checks around agents to catch unsafe or unintended actions.
← all categoriesSEC · category
Privacy & safety
Keeping sensitive data yours while still using powerful cloud models.
6 tools in this category
SEC·02★ 1.2k
power-pwn
A red-team toolkit for probing Microsoft 365 Copilot and Power Platform — for security research, not production.
SEC·03★ 9.0k
Presidio
Detects and redacts personal data — names, emails, IDs — in text and images; a natural partner to PromptMask.
SEC·04★ 127
PromptMask
Local-first privacy layer for LLMs that redacts sensitive data before cloud API calls and restores it in responses.
SEC·05★ 4.3k
Purple Llama
Meta's open trust-and-safety suite for LLM apps. Its notable pieces: Llama Guard (input/output moderation), Prompt Guard (catches prompt injection and jailbreaks), and CodeShield (filters insecure generated code) — plus the CyberSecEval benchmarks for measuring a model's security risk.
SEC·06★ 25k
Strix
Autonomous AI 'hackers' that run your app like a real attacker — finding vulnerabilities, proving them with working proof-of-concepts instead of false positives, and opening fix PRs. Runs from the CLI or in CI on every pull request.