Axiomeer – An open marketplace for AI agents
Hi, I built Axiomeer, an open-source marketplace protocol for AI agents. The idea: instead of hardcoding tool integrations into every agent, agents shop a catalog at runtime, and the marketplace ranks, executes, validates, and audits everything. How it works: - Providers publish products (APIs, datasets, model endpoints) via 10-line JSON manifests - Agents describe what they need in natural language or structured tags - The router scores all options by capability match (70%), latency (20%), cost (10%) with hard constraint filters - The top pick is executed, output is validated (citations…
In plain words
Axiomeer is an open-source marketplace protocol that lets AI agents dynamically discover and use tools, APIs, datasets, and model endpoints at runtime instead of relying on hardcoded integrations. Providers publish services via simple JSON manifests, and agents request capabilities in natural language or structured tags. A router ranks available options by capability match, latency, and cost, then executes the top choice while validating outputs and creating immutable audit logs. The system prioritizes reliability by rejecting low-quality or unverified results rather than generating false answers.
written from the facts on this page · September 2026
From the sources
In the maker’s words, at launch
Hi, I built Axiomeer, an open-source marketplace protocol for AI agents. The idea: instead of hardcoding tool integrations into every agent, agents shop a catalog at runtime, and the marketplace ranks, executes, validates, and audits everything. How it works: - Providers publish products (APIs, datasets, model endpoints) via 10-line JSON manifests - Agents describe what they need in natural language or structured tags - The router scores all options by capability match (70%), latency (20%), cost (10%) with hard constraint filters - The top pick is executed, output is validated (citations required? timestamps?), and evidence quality is assessed deterministically - If the evidence is mock/fake/low-quality, the agent abstains rather than hallucinating - Every execution is logged as an immutable receipt The trust layer is the part I think is missing from existing approaches. MCP standardizes how you connect to a tool server. Axiomeer operates one layer up: which tool, from which provider, and can you trust what came back? Stack: Python, FastAPI, SQLAlchemy, Ollama (local LLM, no API keys). v1 ships with weather providers (Open-Meteo + mocks). The architecture supports any HTTP endpoint that returns structured JSON. Looking for contributors to add real providers across domains (finance, search, docs, code execution). Each provider is ~30 lines + a manifest.
Does the same job
all alternatives →
- MAMeta-agent: self-improving agent harnesses from live tracesApr 2026 · github.com · ▲14
We built meta-agent: an open-source library that automatically and continuously improves agent harnesses from production traces. Point it at an existing agent, a stream of unlabeled production traces, and a small labeled holdout set. An LLM judge scores unlabeled production traces as they stream. A proposer reads failed traces and writes one targeted harness update at a time, such as changes to prompts, hooks, tools, or subagents. The update is kept only if it improves holdout accuracy. On tau-bench v3 airline, meta-agent improved holdout accuracy from 67% to 87%. We open-sourced meta-agent.…
Agentic Operating SystemJul 2026 · ibl.ai · ▲11Open-source AI agent platform — own it, run any model
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
