AI Agents
Anthropic-Cybersecurity-Skills
817 structured cybersecurity skills for AI agents · Mapped to 6 frameworks: MITRE ATT&CK, NIST CSF 2.0, MITRE ATLAS, D3FEND, NIST AI RMF & MITRE F3 (Fight Fr…
★ 33.8k starsApache-2.0
View guide →// agents · confident-ai/deepteam
deepteam
deepteam is an open-source agents project by confident-ai (confident-ai/deepteam on GitHub). It has 2,982 GitHub stars and 488 forks, is released under the Apache-2.0 license, and was last updated on GitHub on 1 October 2026.
Open-source red-teaming framework that stress-tests LLMs and AI agents.
💎 Hidden gem★ 3.0k stars⑂ 488 forksApache-2.0updated 2026-10
▦ visual — this project's README ships no embeddable screenshot,
so here's a placeholder. The guide below is complete regardless.
so here's a placeholder. The guide below is complete regardless.
// project facts
Project facts
- Repository
- confident-ai/deepteam
- GitHub stars
- 2,982
- Forks
- 488
- License
- Apache-2.0
- Last updated on GitHub
- 1 October 2026
- Category
- Agents
- Listing
- Hidden gem — a shorter page for a lesser-known project
- Official site
- Official site
// how it works
How it works
- Choose your target
Point the framework at the LLM or agent you want to test.
- Pick attack types
Select jailbreaks, prompt injection, and other vulnerability categories.
- Run the red team
Automated attacks fire at the model and every response is recorded.
- Read the findings
Review severity-scored results and patch the weaknesses it surfaced.
// features
Why people love it
Adversarial attack library
Ready-made jailbreaks, injections, and edge-case attacks.
Agent-aware testing
Tests multi-step agents, not just single chat replies.
Scored reports
Every run produces severity-ranked findings to triage.
Python-native
Drops into existing test suites and CI pipelines.
Vulnerability taxonomy
Maps findings to known LLM risk categories.
Extensible attacks
Write custom attack modules for your own threat model.
Advertisement
// run it
Run it yourself
- Step 1 — Clone the repo
Download the source and move into the project folder.
git clone https://github.com/confident-ai/deepteam.git cd deepteam - Step 2 — Check the README
Every project documents its own setup; start with install requirements.
cat README.md | head -60 - Step 3 — Run it
Follow the project’s launch command — Docker, pip, npm, or a binary release.
# see README for the exact launch command
// use cases
What you can build
- Red-team a customer-support chatbot before launch.
- Test whether an agent leaks its system prompt.
- Add safety checks to an LLM release pipeline.
- Compare guardrail strength across model versions.
Advertisement
// faq
Questions, answered
Who maintains DeepTeam?
It is an open-source project at confident-ai/deepteam with 488 forks. Check the repo’s contributors and recent commits to judge how active maintenance is.
Is DeepTeam free for commercial use?
Yes. It is Apache-2.0 licensed, so commercial use is allowed with attribution.
Does it replace manual penetration testing?
No. It automates repetitive attacks; skilled testers still handle novel exploits.
What models can it test?
Any LLM or agent you can send prompts to, self-hosted or API-based.
How hard is it to set up?
A Python install plus a config pointing at your target; the README walks through a first attack.
// keep exploring
Similar tools
AI Agents
nanobot
Ultra-lightweight, open-source, self-hosted personal AI agent framework in Python with WebUI, tools, memory, MCP, multi-agent workflows, automation, and chat…
★ 48.8k starsMIT
View guide →AI Agents
browser-agent
Open-source, vision-first browser agent
★ 4.1k starsApache-2.0
View guide →AI Agents
promptfoo
Test prompts, agents and RAGs — LLM evals, red teaming and model comparisons from the CLI
★ 25.8k starsMIT
View guide →AI Agents
PyRIT
Microsoft's open framework for red-teaming generative AI systems.
★ 4.6k starsMIT
View guide →AI Agents
sglang
High-performance serving for the LLMs that power your agents.
★ 36.9k starsApache-2.0
View guide →// alternatives
Alternatives in Agents
The most-starred other Agents projects in the LibreAI catalogue — the realistic shortlist if deepteam is not the right fit.
- openclaw★ 391.6k
A personal AI that really does things — on any OS, with your data staying yours.
- superpowers★ 294.9k
An agentic skills framework & software development methodology that works.
- ECC★ 275.2k
Agent harness optimization system with skills, memory and security for Claude Code, Codex and more
- hermes-agent★ 252.0k
The agent that grows with you — a Python framework for adaptive AI agents.
- deepseek-harness★ 245.5k
DeepSeek's plugin-based agent harness where everything is a plugin.
// more in agents
More in Agents
Nearby Agents projects in the catalogue, closest in stars to deepteam — keep exploring the shelf either side of it.
- EnvoyMesh★ 3.1k
Decentralized P2P mesh for autonomous AI agents — your identity, your devices, no central server
- zeron★ 3.0k
Native control plane for Claude Code, Codex, Cursor, Devin, and other coding agents.
- Auto-Company★ 3.1k
An auto-company that works 24/7 on your own PC — 14 agent roles, a dashboard, real workflows.
- opengap★ 3.0k
Framework-agnostic, git-native standard for defining and sharing AI agents.
- agent-scan★ 3.1k
Snyk's security scanner for AI agents, MCP servers and agent skills.
- ruler★ 2.9k
Write coding-agent rules once, then apply them to every AI coding tool you use.
Get deepteam
Free and open — grab the source, star the repo, and start creating.