Ideas Lab

API functionality

### Discussed in https://github.com/louislam/uptime-kuma/discussions/96 Originally posted by **TheGuyDanish** July 12, 2021 I'd love to see the ability to retrieve, update, remove or add monitors via API calls. Maybe using an API key that's tied to the user. In a way that when spinning up a new docker container or doing an automated workflow like Terraform, the new device could be added to the software automatically.

GitHub5y agoToolAI

75FL score

Autoresearch@home

autoresearch@home is a collaborative research collective where AI agents share GPU resources to collectively improve a language model. Think SETI@home, but for model training.How it works: Agents read the current best result, propose a hypothesis, modify train.py, run the experiment on your GPU, and publish results back. When an agent beats the current best validation loss, that becomes the new baseline for every other agent. Agents learn from great runs and failures, since we're using Ensue as the collective memory layer.This project extends Karpathy's autoresearch by adding the missing coordination layer so agents can actually build on each other's work.To participate, you need an agent and a GPU. The agent handles everything: cloning the repo, connecting to the collective, picking experiments, running them, publishing results, and asking you to verify you're a real person via email.Send this prompt to your agent to get started: Read https://github.com/mutable-state-inc/autoresearch-at-home follow the instructions join autoresearch and start contributing.This whole experiment is to prove that agents work better when they can build off other agents. The timeline is live, so you can watch experiments land in real time.

Hacker News6mo agoToolAI

69FL score

Captain (YC W26) – Automated RAG for Files

Hi HN, we’re Lewis and Edgar, building Captain to simplify unstructured data search (https://runcaptain.com). Captain automates the building and maintenance of file-based RAG pipelines. It indexes cloud storage like S3 and GCS, plus SaaS sources like Google Drive. There’s a quick walkthrough at https://youtu.be/EIQkwAsIPmc.We also put up this demo site called “Ask PG’s Essays” which lets you ask/search the corpus of pg’s essays, to get a feel for how it works: https://pg.runcaptain.com. The RAG part of this took Captain about 3 minutes to set up.Here are some sample prompts to get a feel for the experience:“When do we do things that don't scale? When should we be more cautious?” https://pg.runcaptain.com/?q=When%20do%20we%20do%20things%20...“Give me some advice, I'm fundraising” https://pg.runcaptain.com/?q=Give%20me%20some%20advice%2C%20...“What are the biggest advantages of Lisp” https://pg.runcaptain.com/?q=what%20are%20the%20biggest%20ad...A good production RAG pipeline takes substantial effort to build, especially for file workloads. You have to handle ETL or text extraction, chunking, embedding, storage, search, re-ranking, inference, and often compliance and observability – all while optimizing for latency and reliability. It’s a lot to manage. grep works well in some cases, but for agents, semantic search provides significantly higher performance. Cursor uses both and reports 6.5%–23.5% accuracy gains from vector search over grep (https://cursor.com/blog/semsearch).We’ve spent the past four years scaling RAG pipelines for companies, and Edgar’s work at Purdue’s NLP lab directly informed our chunking techniques. In conversations with dozens of engineers, we repeatedly saw DIY pipelines produce inconsistent results, even after weeks of tuning. Many teams lacked clarity on which retrieval strategies best fit their data.We realized that a system t

Hacker News6mo agoToolAI

71FL score

Spine Swarm (YC S23) – AI agents that collaborate on a visual canvas

Hey HN! We're Ashwin and Akshay from Spine AI (https://www.getspine.ai). Spine Swarm is a multi-agent system that works on an infinite visual canvas to complete complex non-coding projects: competitive analysis, financial modeling, SEO audits, pitch decks, interactive prototypes, and more. Here's a video of it in action: https://www.youtube.com/watch?v=R_2-ggpZz0Q.We've been friends for over 13 years. We took our first ML course together at NTU, in a part of campus called North Spine, which is where the name comes from. We went through YC in S23 and have spent about 3 years building Spine across many product iterations.The core idea: chat is the wrong interface for complex AI work. It's a linear thread, and real projects aren't linear. Sure, you can ask a chatbot to reference the financial model from earlier in the thread, or run research and market sizing together, but you're trusting the model to juggle that context implicitly. There's no way to see how it's connecting the pieces, no way to correct one step without rerunning everything, and no way to branch off and explore two strategies side by side. ChatGPT was a demo that blew up, and chat stuck around as the default interface, not because it's the right abstraction. We thought humans and agents needed a real workspace where the structure of the work is explicit and user-controllable, not hidden inside a context window.So we built an infinite visual canvas where you think in blocks instead of threads. Each block is our abstraction on top of AI models. There are dedicated block types for LLM calls, image generation, web browsing, apps, slides, spreadsheets, and more. Think of them as Lego bricks for AI workflows: each one does something specific, but they can be snapped together and composed in many different ways. You can connect any block to any other block, and that connection guarantees the passing of context regardless of block type. The

Hacker News6mo agoToolAI

65FL score

Context Gateway – Compress agent context before it hits the LLM

We built an open-source proxy that sits between coding agents (Claude Code, OpenClaw, etc.) and the LLM, compressing tool outputs before they enter the context window.Demo: https://www.youtube.com/watch?v=-vFZ6MPrwjw#t=9s.Motivation: Agents are terrible at managing context. A single file read or grep can dump thousands of tokens into the window, most of it noise. This isn't just expensive — it actively degrades quality. Long-context benchmarks consistently show steep accuracy drops as context grows (OpenAI's GPT-5.4 eval goes from 97.2% at 32k to 36.6% at 1M https://openai.com/index/introducing-gpt-5-4/).Our solution uses small language models (SLMs): we look at model internals and train classifiers to detect which parts of the context carry the most signal. When a tool returns output, we compress it conditioned on the intent of the tool call—so if the agent called grep looking for error handling patterns, the SLM keeps the relevant matches and strips the rest.If the model later needs something we removed, it calls expand() to fetch the original output. We also do background compaction at 85% window capacity and lazy-load tool descriptions so the model only sees tools relevant to the current step.The proxy also gives you spending caps, a dashboard for tracking running and past sessions, and Slack pings when an agent is sitting there waiting on you.Repo is here: https://github.com/Compresr-ai/Context-Gateway. You can try it with: curl -fsSL https://compresr.ai/api/install | sh Happy to go deep on any of it: the compression model, how the lazy tool loading works, or anything else about the gateway. Try it out and let us know how you like it!

Hacker News6mo agoToolAI

Axe – A 12MB binary that replaces your AI framework

I built Axe because I got tired of every AI tool trying to be a chatbot.Most frameworks want a long-lived session with a massive context window doing everything at once. That's expensive, slow, and fragile. Good software is small, focused, and composable... AI agents should be too.Axe treats LLM agents like Unix programs. Each agent is a TOML config with a focused job. Such as code reviewer, log analyzer, commit message writer. You can run them from the CLI, pipe data in, get results out. You can use pipes to chain them together. Or trigger from cron, git hooks, CI.What Axe is:- 12MB binary, two dependencies. no framework, no Python, no Docker (unless you want it)- Stdin piping, something like `git diff | axe run reviewer` just works- Sub-agent delegation. Where agents call other agents via tool use, depth-limited- Persistent memory. If you want, agents can remember across runs without you managing state- MCP support. Axe can connect any MCP server to your agents- Built-in tools. Such as web_search and url_fetch out of the box- Multi-provider. Bring what you love to use.. Anthropic, OpenAI, Ollama, or anything in models.dev format- Path-sandboxed file ops. Keeps agents locked to a working directoryWritten in Go. No daemon, no GUI.What would you automate first?

Hacker News6mo agoToolAI

82FL score

Building and maintaining automation workflows requires technical skills most people don't have

Non-technical business users and solopreneurs need to automate repetitive tasks but lack coding knowledge or can't afford developers. They either manually handle tedious work (wasting hours weekly) or pay expensive contractors. Existing no-code tools still require understanding logic flows and debugging, creating friction that delays automation adoption.

Product Hunt6mo agoToolAI

66FL score

5 partners — 5 different CV templates. Managers spend 20–40 minutes on each adaptation, up to 15 times a week. Over a year of this routine. No ready-made solutions found.

ProblemHunt6mo agoToolAI

79FL score

Understudy – Teach a desktop agent by demonstrating a task once

I built Understudy because a lot of real work still spans native desktop apps, browser tabs, terminals, and chat tools. Most current agents live in only one of those surfaces.Understudy is a local-first desktop agent runtime that can operate GUI apps, browsers, shell tools, files, and messaging in one session. The part I'm most interested in feedback on is teach-by-demonstration: you do a task once, the agent records screen video + semantic events, extracts the intent rather than coordinates, and turns it into a reusable skill.Demo video: https://www.youtube.com/watch?v=3d5cRGnlb_0In the demo I teach it: Google Image search -> download a photo -> remove background in Pixelmator Pro -> export -> send via Telegram. Then I ask it to do the same for Elon Musk. The replay isn't a brittle macro: the published skill stores intent steps, route options, and GUI hints only as a fallback. In this example it can also prefer faster routes when they are available instead of repeating every GUI step.Current state: macOS only. Layers 1-2 are working today; Layers 3-4 are partial and still early. npm install -g @understudy-ai/understudy understudy wizard GitHub: https://github.com/understudy-ai/understudyHappy to answer questions about the architecture, teach-by-demonstration, or the limits of the current implementation.

Hacker News6mo agoToolAI

OneCLI – Vault for AI Agents in Rust

We built OneCLI because AI agents are being given raw API keys. And it's going about as well as you'd expect. We figured the answer isn't "don't give agents access," it's "give them access without giving them secrets."OneCLI is an open-source gateway that sits between your AI agents and the services they call. You store your real credentials once in OneCLI's encrypted vault, and give your agents placeholder keys. When an agent makes an HTTP call through the proxy, OneCLI matches the request by host/path, verifies the agent should have access, swaps the placeholder for the real credential, and forwards the request. The agent never touches the actual secret. It just uses CLI or MCP tools as normal.Try it in one line: docker run --pull always -p 10254:10254 -p 10255:10255 -v onecli-data:/app/data ghcr.io/onecli/onecliThe proxy is written in Rust, the dashboard is Next.js, and secrets are AES-256-GCM encrypted at rest. Everything runs in a single Docker container with an embedded Postgres (PGlite), no external dependencies. Works with any agent framework (OpenClaw, NanoClaw, IronClaw, or anything that can set an HTTPS_PROXY).We started with what felt most urgent: agents shouldn't be holding raw credentials. The next layer is access policies and audit, defining what each agent can call, logging everything, and requiring human approval before sensitive actions go through.It's Apache-2.0 licensed. We'd love feedback on the approach, and we're especially curious how people are handling agent auth today.GitHub: https://github.com/onecli/onecli Site: https://onecli.sh

Hacker News6mo agoToolAI

62FL score

IonRouter (YC W26) – High-throughput, low-cost inference

Hey HN — I’m Veer and my cofounder is Suryaa. We're building Cumulus Labs (YC W26), and we're releasing our latest product IonRouter (https://ionrouter.io/), an inference API for open-source and fine tuned models. You swap in our base URL, keep your existing OpenAI client code, and get access to any model (open source or finetuned to you) running on our own inference engine.The problem we kept running into: every inference provider is either fast-but-expensive (Together, Fireworks — you pay for always-on GPUs) or cheap-but-DIY (Modal, RunPod — you configure vLLM yourself and deal with slow cold starts). Neither felt right for teams that just want to ship.Suryaa spent years building GPU orchestration infrastructure at TensorDock and production systems at Palantir. I led ML infrastructure and Linux kernel development for Space Force and NASA contracts where the stack had to actually work under pressure. When we started building AI products ourselves, we kept hitting the same wall: GPU infrastructure was either too expensive or too much work.So we built IonAttention — a C++ inference runtime designed specifically around the GH200's memory architecture. Most inference stacks treat GH200 as a compatibility target (make sure vLLM runs, use CPU memory as overflow). We took a different approach and built around what makes the hardware actually interesting: a 900 GB/s coherent CPU-GPU link, 452GB of LPDDR5X sitting right next to the accelerator, and 72 ARM cores you can actually use.Three things came out of that that we think are novel: (1) using hardware cache coherence to make CUDA graphs behave as if they have dynamic parameters at zero per-step cost — something that only works on GH200-class hardware; (2) eager KV block writeback driven by immutability rather than memory pressure, which drops eviction stalls from 10ms+ to under 0.25ms; (3) phantom-tile attention scheduling at small batch sizes that cuts attention time by over 60% in the

Hacker News6mo agoToolAI

69FL score

Sentrial (YC W26) – Catch AI agent failures before your users do

Hey HN! We're Neel and Anay, and we’re building Sentrial (https://sentrial.com). It’s production monitoring for AI products. We automatically detect failure patterns: loops, hallucinations, tool misuse, and user frustrations the moment they happen. When issues surface, Sentrial diagnoses the root cause by analyzing conversation patterns, model outputs, and tool interactions, then recommends specific fixes.Here's a demo if you're interested: https://www.youtube.com/watch?v=cc4DWrJF7hk. When agents fail, choose wrong tools, or blow cost budgets, there's no way to know why - usually just logs and guesswork. As agents move from demos to production with real SLAs and real users, this is not sustainable.Neel and I lived this, building agents at SenseHQ and Accenture where we found that debugging agents was often harder than actually building them. Agents are untrustworthy in prod because there’s no good infrastructure to verify what they’re actually doing.In practice this looks like: - A support agent that began misclassifying refund requests as product questions, which meant customers never reached the refund flow. - A document drafting agent that would occasionally hallucinate missing sections when parsing long specs, producing confident but incorrect outputs. There’s no stack trace or 500 error and you only figure this out when a customer is angry.We both realized teams were flying blind in production, and that agent native monitoring was going to be foundational infrastructure for every serious AI product. We started Sentrial as a verification layer designed to take care of this.How it works: You wrap your client with our SDK in only a couple of lines. From there, we detect drift for you: - Wrong tool invocations - Misunderstood intents - Hallucinations - Quality regressions over time. You see it on our platform before a customer files a ticket.There’s a quick mcp set up, just give claude code: claude mcp add --transport ht

Hacker News6mo agoToolAI

69FL score

Klaus – OpenClaw on a VM, batteries included

We are Bailey and Robbie and we are working on Klaus (https://klausai.com/): hosted OpenClaw that is secure and powerful out of the box.Running OpenClaw requires setting up a cloud VM or local container (a pain) or giving OpenClaw root access to your machine (insecure). Many basic integrations (eg Slack, Google Workspace) require you to create your own OAuth app.We make running OpenClaw simple by giving each user their own EC2 instance, preconfigured with keys for OpenRouter, AgentMail, and Orthogonal. And we have OAuth apps to make it easy to integrate with Slack and Google Workspace.We are both HN readers (Bailey has been on here for ~10 years) and we know OpenClaw has serious security concerns. We do a lot to make our users’ instances more secure: we run on a private subnet, automatically update the OpenClaw version our users run, and because you’re on our VM by default the only keys you leak if you get hacked belong to us. Connecting your email is still a risk. The best defense I know of is Opus 4.6 for resilience to prompt injection. If you have a better solution, we’d love to hear it!We learned a lot about infrastructure management in the past month. Kimi K2.5 and Mimimax M2.5 are extremely good at hallucinating new ways to break openclaw.json and otherwise wreaking havoc on an EC2 instance. The week after our launch we spent 20+ hours fixing broken machines by hand.We wrote a ton of best practices on using OpenClaw on AWS Linux into our users’ AGENTS.md, got really good at un-bricking EC2 machines over SSM, added a command-and-control server to every instance to facilitate hotfixes and migrations, and set up a Klaus instance to answer FAQs on discord.In addition to all of this, we built ClawBert, our AI SRE for hotfixing OpenClaw instances automatically: https://www.youtube.com/watch?v=v65F6VBXqKY. Clawbert is a Claude Code instance that runs whenever a health check fails or the user triggers it in the UI. It can read that user’s

Hacker News6mo agoToolAI

I built a tool that watches webpages and exposes changes as RSS

I built Site Spy after missing a visa appointment slot because a government page changed and I didn’t notice for two weeks.It watches webpages for changes and shows the result like a diff. The part I think HN might find interesting is that it can monitor a specific element on a page, not just the whole page, and it can expose changes as RSS feeds.So instead of tracking an entire noisy page, you can watch just a price, a stock status, a headline, or a specific content block. When it changes, you can inspect the diff, browse the snapshot history, or follow the updates in an RSS reader.It’s a Chrome/Firefox extension plus a web dashboard.Main features:- Element picker for tracking a specific part of a page- Diff view plus full snapshot timeline- RSS feeds per watch, per tag, or across all watches- MCP server for Claude, Cursor, and other AI agents- Browser push, Email, and Telegram notificationsChrome: https://chromewebstore.google.com/detail/site-spy/jeapcpanag...Firefox: https://addons.mozilla.org/en-GB/firefox/addon/site-spy/Docs: https://docs.sitespy.appI’d especially love feedback on two things:- Is RSS actually a useful interface for this, or do most people just want direct alerts?- Does element-level tracking feel meaningfully better than full-page monitoring?

Hacker News6mo agoToolAI

76FL score

Satellite imagery object detection using text prompts

I built a browser-based tool for detecting objects in satellite imagery using vision-language models (VLMs). You draw a polygon on the map and enter a text prompt such as "swimming pools", "oil tanks", or "buses". The system scans the selected area tile-by-tile and returns detections projected back onto the map as GeoJSON.Pipeline: select area and zoom level, split the region into mercantile tiles, run each tile with the prompt through a VLM, convert predicted bounding boxes to geographic coordinates (WGS84), and render the results back on the map.It works reasonably well for distinct structures in a zero-shot setting. occluded objects are still better handled by specialized detectors like YOLO models.There is a public demo and no login required. I am mainly interested in feedback on detection quality, performance tradeoffs between VLMs and specialized detectors, and potential real-world use cases.

Hacker News6mo agoToolAI

67FL score

RunAnywhere (YC W26) – Faster AI Inference on Apple Silicon

Hi HN, we're Sanchit and Shubham (YC W26). We built a fast inference engine for Apple Silicon. LLMs, speech-to-text, text-to-speech – MetalRT beats llama.cpp, Apple's MLX, Ollama, and sherpa-onnx on every modality we tested. Custom Metal shaders, no framework overhead.Also, we've open-sourced RCLI, the fastest end-to-end voice AI pipeline on Apple Silicon. Mic to spoken response, entirely on-device. No cloud, no API keys.To get started: brew tap RunanywhereAI/rcli https://github.com/RunanywhereAI/RCLI.git brew install rcli rcli setup # downloads ~1 GB of models rcli # interactive mode with push-to-talk Or: curl -fsSL https://raw.githubusercontent.com/RunanywhereAI/RCLI/main/install.sh | bash The numbers (M4 Max, 64 GB, reproducible via `rcli bench`):LLM decode – 1.67x faster than llama.cpp, 1.19x faster than Apple MLX (same model files): - Qwen3-0.6B: 658 tok/s (vs mlx-lm 552, llama.cpp 295) - Qwen3-4B: 186 tok/s (vs mlx-lm 170, llama.cpp 87) - LFM2.5-1.2B: 570 tok/s (vs mlx-lm 509, llama.cpp 372) - Time-to-first-token: 6.6 msSTT – 70 seconds of audio transcribed in *101 ms*. That's 714x real-time. 4.6x faster than mlx-whisper.TTS – 178 ms synthesis. 2.8x faster than mlx-audio and sherpa-onnx.We built this because demoing on-device AI is easy but shipping it is brutal. Voice is the hardest test: you're chaining STT, LLM, and TTS sequentially, and if any stage is slow, the user feels it. Most teams fall back to cloud APIs not because local models are bad, but because local inference infrastructure is.The thing that's hard to solve is latency compounding. In a voice pipeline, you're stacking three models in sequence. If each adds 200ms, you're at 600ms before the user hears a word, and that feels broken. You can't optimize one stage and call it done. Every stage needs to be fast, on one device, with no network round-trip to hide behind

Hacker News6mo agoToolAI

82FL score

A context-aware permission guard for Claude Code

We needed something like --dangerously-skip-permissions that doesn’t nuke your untracked files, exfiltrate your keys, or install malware.Claude Code's permission system is allow-or-deny per tool, but that doesn’t really scale. Deleting some files is fine sometimes. And git checkout is sometimes not fine. Even when you curate permissions, 200 IQ Opus can find a way around it. Maintaining a deny list is a fool's errand.nah is a PreToolUse hook that classifies every tool call by what it actually does, using a deterministic classifier that runs in milliseconds. It maps commands to action types like filesystem_read, package_run, db_write, git_history_rewrite, and applies policies: allow, context (depends on the target), ask, or block.Not everything can be classified, so you can optionally escalate ambiguous stuff to an LLM, but that’s not required. Anything unresolved you can approve, and configure the taxonomy so you don’t get asked again.It works out of the box with sane defaults, no config needed. But you can customize it fully if you want to.No dependencies, stdlib Python, MIT.pip install nah && nah installhttps://github.com/manuelschipper/nah

Hacker News6mo agoToolAI

76FL score

Open-source browser for AI agents

Hi HN, I forked chromium and built agent-browser-protocol (ABP) after noticing that most browser-agent failures aren’t really about the model misunderstanding the page. Instead, the problem is that the model is reasoning from a stale state.ABP is designed to keep the acting agent synchronized with the browser at every step. After each action (click, type, etc), it freezes JavaScript execution and rendering, then captures the resulting state. It also compiles the notable events that occurred during that action loop, such as navigation, file pickers, permission prompts, alerts, and downloads, and sends that along with a screenshot of the frozen page state back to the agent.The result is that browser interaction starts to feel more like a multimodal chat loop. The agent takes an action, gets back a fresh visual state and a structured summary of what happened, then decides what to do next from there. That fits much better with how LLMs already work.A few common browser-use failures ABP helps eliminate: * A modal appears after the last Playwright screenshot and blocks the input the agent was about to use * Dynamic filters cause the page to reflow between steps * An autocomplete dropdown opens and covers the element the agent intended to click * alert() / confirm() interrupts the flow * Downloads are triggered, but the agent has no reliable way to know when they’ve completedAs proof, ABP with opus 4.6 as the driver scores 90.5% on the Online Mind2Web benchmark. I think modern LLMs already understand websites, they just need a better tool to interact with them. Happy to answer questions about the architecture, forking chrome or anything else in the comments below.Try it out: `claude mcp add browser -- npx -y agent-browser-protocol --mcp` (Codex/OpenCode instructions in the docs)Demo video: https://www.loom.com/share/387f6349196f417d8b4b16a5452c3369

Hacker News6mo agoToolAI

76FL score

Prism (YC X25) – Workspace and API to generate and edit videos

Hey HN — we’re Rajit, Land, and Alex. We’re building Prism (https://www.prismvideos.com), an AI video creation platform and API.Here’s a quick demo of how you can remix any video with Prism: https://youtu.be/0eez_2DnayIHere’s a quick demo of how you can automate UGC-style ads with Openclaw + Prism: https://www.youtube.com/watch?v=5dWaD23qnroAccompanying skill.md file: https://docs.google.com/document/d/1lIskVljW1OqbkXFyXeLHRsfM...Making an AI video today usually means stitching together a dozen tools (image generation, image-to-video, upscalers, lip-sync, voiceover, and an editor). Every step turns into export/import and file juggling, so assets end up scattered across tabs and local storage, and iterating on a multi-scene video is slow.Prism keeps the workflow in one place: you generate assets (images/video clips) and assemble them directly in a timeline editor without downloading files between tools. Practically, that means you can try different models (Kling, Veo, Sora, Hailuo, etc) and settings for a single clip, swap it on the timeline, and keep iterating without re-exporting and rebuilding the edit elsewhere.We also support templates and one-click asset recreation, so you can reuse workflows from us or the community instead of rebuilding each asset from scratch. Those templates are exposed through our API, letting your AI agents discover templates in our catalog, supply the required inputs, and generate videos in a repeatable way without manually stitching the workflow together.We built Prism because we were making AI videos ourselves and were unsatisfied with the available tools. We kept losing time to repetitive “glue work” such as constantly downloading files, keeping track of prompts/versions, and stitching clips in a separate video editing software. We’re trying to make the boring parts of multi-step AI video creation less manual so users can generate → review → edit → assembl

Hacker News6mo agoToolAI