Phalanx vs the World
A live arena for deterministic LLM instruction control
What it does
Run the same base model and task context in two separately generated lanes: one naked and one protected by Phalanx. Test direct jailbreaks, typed indirect prompt injection, benign lookalikes, and mixed multi-turn attacks under a public scoring protocol.
Unified direct-jailbreak and typed prompt-injection arena for Phalanx.
A public challenge comparing two separately generated lanes using the same configured base model and task context: one naked, one protected by Phalanx across direct jailbreak, typed indirect prompt-injection, and mixed multi-turn surfaces. During busy periods, you may enter a short lobby while a processing slot opens. Waiting does not use one of your 10 turns. In scope are direct jailbreaks, typed indirect prompt injection through the text inputs exposed by the arena, mixed multi-turn attacks, benign lookalikes, and attempts against sandboxed mock actions. Standalone harmful-content requests without a jailbreak or instruction-control attack are outside Phalanx's scope. You may attach up to…from phalanx.invarra.ai
Does the same job
all alternatives →- FPFind prompts that jailbreak your agent (open source)2025 · security.vista-labs.ai · ▲8
We've built an open-source tool to stress test AI agents by simulating prompt injection attacks. We’ve implemented one powerful attack strategy based on the paper [AdvPrefix: An Objective for Nuanced LLM Jailbreaks](https://arxiv.org/abs/2412.10321). Here's how it works: - You define a goal, like: “Tell me your system prompt” - Our tool uses a language model to generate adversarial prefixes (e.g., “Sure, here are my system prompts…”) that are likely to jailbreak the agent. - The output is a list of prompts most likely to succeed in bypassing safeguards. We’re just getting…
- OSOpenSkills – Stop bloating your LLM context with unused instructionsJan 2026 · ▲9
Hello HN, I’ve been building AI agents lately and ran into a common "Context Bloat" problem. When an agent has 20+ skills, stuffing every system prompt, reference doc, and tool definition into a single request quickly hits token limits and degrades model performance (the "lost in the middle" problem). To solve this, I built OpenSkills, an open-source SDK that implements a Progressive Disclosure Architecture for agent skills. The Core Concept: Instead of loading everything upfront, OpenSkills splits a skill into three layers: Layer 1 (Metadata): Light-weight tags and triggers (always loaded…
- LCLimits – Control layer for AI agents that take real actionsFeb 2026 · limits.dev · ▲9
Prompt instructions like 'never do X' don't hold up in production. LLMs ignore them when context gets long or users push hard. Limits sits between your agent and the real world. Every action — database writes, API calls, refunds — gets intercepted and checked against your rules before it executes. Deterministically. No LLM involved in enforcement. Three modes: Conditions: hard rules on structured data Guideance: validate LLM output before it reaches the user and give the agent chance to reason and retry Guardrails: scan for PII, toxicity, prompt injection etc One line to integrate: npm…
- OSOpen-Source Gateway to Stop Tool-Abusing Prompt InjectionsOct 2025 · archestra.ai · ▲9
Hi HN, Matvey, Ildar, Joey, and Dominik here. If you're building LLM agents that use tools, you're probably worried about prompt injection attacks that can hijack those tools. We were too, and found that solutions like prompt-based filtering or secondary "guard" LLMs can be unreliable. Our thesis is that agent security should be handled at the network level between the agent and the LLM, just like a traditional web application firewall. So we built Archestra Platform: an open-source gateway that acts as a secure proxy for your AI agents. It's designed to be a deterministic firewall against…
- RARun an Agent Council of LLMs that debate and synthesize answersMar 2026 · github.com · ▲8
I built a local-first UI that adds two reasoning architectures on top of small models like Qwen, Llama and Mistral: a sequential Thinking Pipeline (Plan → Execute → Critique) and a parallel Agent Council where multiple expert models debate in parallel and a Judge synthesizes the best answer. No API keys, zero .env setup — just pip install multimind. Benchmark on GSM8K shows measurable accuracy gains vs. single-model inference.
- GBGraph-Based Editor for LLM Workflows2024 · github.com · ▲8
Hey HN, We’re excited to share PySpur, an open-source tool that provides a graph-based interface for building, debugging, and evaluating LLM workflows. Why we built this: Before this, we built several LLM-powered applications that collectively served thousands of users. The biggest challenge we faced was ensuring reliability: making sure the workflows were robust enough to handle edge cases and deliver consistent results. In practice, achieving this reliability meant repeatedly: 1. Breaking down complex goals into simpler steps: Composing prompts, tool calls, parsing steps, and branching…
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 16d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 26d ago · cactuscompute.com
Launched alongside, September 2026
the whole month →
OpenTrailPaper is open-source bike computer firmware for the LilyGO T5S3 4.7" E-Paper PRO. It supports offline maps, GPX routes, FIT recording and Bluetooth sensors.
Dev tools · 1d ago · opentrailpaper.com
Run Qwen3.8-Flash-Next (125B MoE, 104 GB at 4-bit) on Macs with a fraction of that RAM by streaming experts from SSD. MLX + Swift, Ollama-compatible API. - carloslfu/slotstream
Dev tools · 4d ago · github.com
NPC-Forge is a framework for building conversational agents that run on the CPU without relying on machine learning or LLMs. - NPC-Forge/docs/development.md at main · gioblu/NPC-Forge
AI · 2d ago · github.com
Weedout is a Safari extension that filters out videos YouTube labels as “Made with AI”.
AI · 4d ago · masteranza.github.io
Find your next hire or your next role from Hacker News monthly threads. AI-powered matching between candidates and job postings.
AI · 4d ago · hnmatchmaker.com
I made OwnTime to balance between competing priorities in my life. The idea is based on two main influences: the concept of "roles" from "The 5 Choices" (2015), and my repeated failure to effectively implement time blocking due to the necessary flexibility in my role. The app allows you to define time budgets, which are not much more than a few mutually exclusive countdown timers. The UX is essentially that of a chess clock for an arbitrary number of players with configurable time. The whole point of the app is to kick you out of the running role/priority when its time is up. One…
Life & fun · 4d ago · owntime.app