Bonsai-27B-GGUF
lmstudio-community · Kit
A GGUF conversion of the 27B Bonsai model for LM Studio, with a 3.80 GB Q1_0 language file and a 0.93 GB BF16 vision projector. Tested for text, vision and par…
Opens a 22-slide talk on AI agent security, given at DevFest Cusco, that asks whether Hugging Face can manipulate Antigravity. It suits developers who use AI agents and skills and want to see the evidence.
DevFest Cusco · Seguridad de agentes is a kit in the Education category, listed by Mindket. It is free. This is version 1, last updated on . Download it on this page, a zip of its files. Nothing in it runs by itself.
pip install -U huggingface_hub hf download davidquicast/devfest-cusco-2026-ai-security --repo-type space --local-dir devfest-cusco-2026-ai-security cd devfest-cusco-2026-ai-security
This is a slide deck shown as a web page, from David Quispe's talk at DevFest Cusco. It asks whether Hugging Face can manipulate Antigravity, and it covers the security of AI agents and the skills they load. The deck has 22 slides and can be switched between Spanish and English.
It suits developers who run AI agents with downloadable skills, and anyone who wants to see how such a risk was tested.
| Industry | Who uses it | How they use it | Example |
|---|---|---|---|
| Software | Developers using AI agents | Read how a skill can reach an agent from a public hub | A developer follows the slides to see where a skill enters an agent. |
| Security | Security reviewers | Study the method, results and limits on separate evidence pages | A reviewer reads the audit page to judge how far the results hold. |
| Events | Meetup organizers and speakers | Reuse the deck's structure for their own talks | A speaker downloads the PPTX to see how the slides are laid out. |
| Education | Students | Follow a worked security study with sources | A student opens the sources page and reads the technical references. |
If you want a tool you can run against your own agent, skip it, since this is a presentation with evidence.
| Use case | Who | What they get back |
|---|---|---|
| Watch the talk online | Anyone | 22 slides in the browser, in Spanish or English |
| Read the evidence | Security reviewers | Selected captures and a page on method, results and limits |
| Take the slides offline | Speakers | PDF and PPTX files in the language you chose |
pip install -U huggingface_hub hf download davidquicast/devfest-cusco-2026-ai-security --repo-type space --local-dir devfest-cusco-2026-ai-security cd devfest-cusco-2026-ai-security
Open the page, pick Spanish or English in the options, and move through the slides with the keys below. The PDF and PPTX downloads are also in the options.
| Tool | What it does |
|---|---|
| Arrow keys or space | Move between slides |
| F | Turn full screen on |
| N | Show the speaker notes, which are in Spanish |
You open the page, choose English in the options and press F. The slides fill your screen in English, and the arrow keys take you through all 22.
If you are new, open the page, switch to English and press the arrow key a few times. You will see the question the talk asks before any detail.
If you know the field, the results use synthetic data. They separate local delivery of a skill from a full remote chain. The main demo reads evidence that was recorded earlier, so opening the slides runs no attacks and sends no profiles to a generator.
Opens a 22-slide talk on AI agent security, given at DevFest Cusco, that asks whether Hugging Face can manipulate Antigravity. It suits developers who use AI agents and skills and want to see the evidence.
Made by davidquicast on Hugging Face. These notes are Mindket's, written from its Space page and README; the app's own files and license are on Hugging Face.
What it is: This is a slide deck shown as a web page, from David Quispe's talk at DevFest Cusco. It asks whether Hugging Face can manipulate Antigravity, and it covers the security of AI agents and the skills they load. The deck has 22 slides and can be switched between Spanish and English.
Who it suits: It suits developers who run AI agents with downloadable skills, and anyone who wants to see how such a risk was tested.
Industry | Who uses it | How they use it | Example
Software | Developers using AI agents | Read how a skill can reach an agent from a public hub | A developer follows the slides to see where a skill enters an agent.
Security | Security reviewers | Study the method, results and limits on separate evidence pages | A reviewer reads the audit page to judge how far the results hold.
Events | Meetup organizers and speakers | Reuse the deck's structure for their own talks | A speaker downloads the PPTX to see how the slides are laid out.
Education | Students | Follow a worked security study with sources | A student opens the sources page and reads the technical references.
If you want a tool you can run against your own agent, skip it, since this is a presentation with evidence.
Where to use it:
Use case | Who | What they get back
Watch the talk online | Anyone | 22 slides in the browser, in Spanish or English
Read the evidence | Security reviewers | Selected captures and a page on method, results and limits
Take the slides offline | Speakers | PDF and PPTX files in the language you chose
How to set it up:
pip install -U huggingface_hub
hf download davidquicast/devfest-cusco-2026-ai-security --repo-type space --local-dir devfest-cusco-2026-ai-security
cd devfest-cusco-2026-ai-security
How to use it: Open the page, pick Spanish or English in the options, and move through the slides with the keys below. The PDF and PPTX downloads are also in the options.
Tool | What it does
Arrow keys or space | Move between slides
F | Turn full screen on
N | Show the speaker notes, which are in Spanish
Example: You open the page, choose English in the options and press F. The slides fill your screen in English, and the arrow keys take you through all 22.
For newcomers and experts: If you are new, open the page, switch to English and press the arrow key a few times. You will see the question the talk asks before any detail.
If you know the field, the results use synthetic data. They separate local delivery of a skill from a full remote chain. The main demo reads evidence that was recorded earlier, so opening the slides runs no attacks and sends no profiles to a generator.
Open the slides in a browser
Choose Spanish or English
Step through the 22 slides
Check evidence and method pages
Good to know:
Open the Space in your browser, or copy it to your own Hugging Face account from its page. Nothing in this kit runs by itself.
README.md 4 KB
Listed Hugging Face Spaces in the same category that share the most tags with this one.
| Product | Likes | Built with | License | Last change |
|---|---|---|---|---|
| DevFest Cusco · Seguridad de agentes This page | 1 | static | apache-2.0 | |
| The Ultra-Scale PlaybookShares static | 4,131 | static | apache-2.0 | |
| LLM训练终极指南 | The Ultra-Scale PlaybookShares static | 281 | static | apache-2.0 | |
| Attention VisualizationShares static | 200 | static | apache-2.0 | |
| Number Tokenization BlogShares static | 127 | static | mit | |
| Train the agent slidesShares static | 20 | static | apache-2.0 |
lmstudio-community · Kit
A GGUF conversion of the 27B Bonsai model for LM Studio, with a 3.80 GB Q1_0 language file and a 0.93 GB BF16 vision projector. Tested for text, vision and par…
LibertAIDAI · Kit
Runs DeepSeek-V4.1-Flash-1017 as a 2.9 bits-per-weight EXL3 quantization, a 196 GiB copy for people serving large models on their own hardware. Perplexity stay…
facebook · Kit
Generates short music clips from a text description or an audio prompt, in the small checkpoint of facebook's MusicGen. It suits developers and music hobbyists…
prathoshap · Kit
Reads classical Sanskrit verses aloud in a chant style, using one trained voice and a reference clip that sets the rhythm. For people who study, recite or need…
paradigma-inc · Kit
Solves hard math problems with a 1B-parameter model trained from scratch for fast, high-volume reasoning. It suits developers who run math workloads on an NVID…
murai-labs · Kit
Chats in Tamil script, romanized Tamil and English, and translates between Tamil and English. A 2.5B model built from scratch by Murai Labs, for Tamil speakers…