Kontext.dev – Runtime Credentials for Agents
Every AI agent that does something useful - opening a PR, posting in Slack, updating a ticket - needs to call an API on behalf of a user. That means OAuth. Authorization flows, token storage, refresh logic, per-user credential isolation. Today, most teams solve this with a long-lived API key in an .env file, shared across every user and every session. As everyone in an organization becomes a software engineer - whether they know it or not - you can't expect each of them to roll their own OAuth. Secure credential handling needs to be baked into the infrastructure. The core issue: agents are a…
What it does
In the maker’s words, at launch
Every AI agent that does something useful - opening a PR, posting in Slack, updating a ticket - needs to call an API on behalf of a user. That means OAuth. Authorization flows, token storage, refresh logic, per-user credential isolation. Today, most teams solve this with a long-lived API key in an .env file, shared across every user and every session. As everyone in an organization becomes a software engineer - whether they know it or not - you can't expect each of them to roll their own OAuth. Secure credential handling needs to be baked into the infrastructure. The core issue: agents are a new kind of principal. They're not users. They're not service accounts. They act with delegated authority on someone else's behalf. Existing identity systems have no model for this. Kontext does. Kontext validates the user and the agent, and returns a credential that lets agents act on behalf of their user - scoped, short-lived, and revocable. No secrets in your codebase, and action is traced back to the human who approved it. We are thrilled to be launching our SDK, let us know what you think!
Does the same job
all alternatives →
MonoCloud for StartupsJul 2026 · monocloud.com · ▲167One identity layer for your customers, APIs, and agents
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 · 17d 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 · 27d ago · cactuscompute.com
Launched alongside, March 2026
the whole month →
Switch from ChatGPT to Claude with import memory feature
AI · Mar 2026 · claude.com
