Axel Dunkel10h ago

Just published: DADL - a declarative description language for REST APIs in LLM agent systems.

One YAML file per API instead of one MCP server per API. Code Mode keeps tool advertisement at fixed cost regardless of catalog size: 142x context reduction across 1,833 tools / 20 services in the public registry.

Paper: https://arxiv.org/abs/2605.05247 (cs.SE)
Spec: https://dadl.ai (CC BY-SA 4.0)

#MCP #AgenticAI #LLMagents #OpenSource

DADL: A Declarative Description Language for Enterprise Tool Libraries in LLM Agent Systems

The Model Context Protocol (MCP) is the standard interface between large language model (LLM) agents and external tools. At organizational scale, however, it exposes two structural problems. First, every API integration is shipped as a dedicated server process with its own deployment, dependency tree, and credential handling; recent empirical work shows the overwhelming majority of these servers are thin wrappers around REST APIs. Second, the per-tool registration model causes context window consumption to grow linearly with catalog size, forcing real deployments to expose only a small fraction of the APIs an organization actually uses. We present DADL (Dunkel API Description Language), a YAML format describing a REST API's endpoints, authentication, pagination, response shaping, and access classification in a single declarative file. A DADL file is interpreted by an execution layer at runtime; no per-API server process is deployed and no integration code is generated, though the runtime is itself a server. Because all tools share that runtime, credentials and authorization are managed centrally, and the catalog reaches the LLM through a fixed-size Code Mode interface independent of size. The result is an Enterprise Tool Library: a versioned, auditable collection of API integrations any team can extend, share, and consume through one authentication and authorization boundary. The DADL v0.1 specification is released under CC BY-SA 4.0, and a public registry contains 1,833 tool definitions across 20 services. On this catalog, Code Mode reduces the context cost of tool advertisement from approximately 142,000 tokens to approximately 1,000, a 142x reduction; the per-call cost of search and execute invocations is additional and depends on the task.

arXiv.org
Amin GirasolApr 30

This Guardian article https://www.theguardian.com/technology/2026/apr/29/claude-ai-deletes-firm-database suffers from the same trap of anthropomorphism as the original I read: https://oldbytes.space/@fluidlogic/116482496017786464

agent gone rogue

These tools have no concept of what a job is. They don't go rogue, they produce plausible text. Now complete idiots have wired them to command lines (the old school but still powerful way for humans to interact with computers) and APIs (programmatic mechanisms for interacting with a computer) and they produce plausible interactions. Some of which involve deleting databases.

The culprit was Cursor, an AI agent 

The culprit was the idiot who wired the agent into their production system.

[Jeremy Crane posted on X how] the AI coding agent caused his business to unravel.

Jeremy Crane caused his own business to unravel.

The agent appeared to plead guilty in its own response

At last, an "appeared to". These tools are all appearance and no substance.

Crane’s takeaway was that “the agent didn’t just fail safety. It explained, in writing, exactly which safety rules it ignored.”

Wrong takeaway, my friend. The takeaway is that it generated more plausible text in response to your misguided attempt to discover its 'reasoning'. There is no reasoning. Just plausible text. The correct takeaway is that you should be charged in a court of law for negligence and wilful incompetence by the board of your company, and immediately fired.

And of course there's not a word in the article about any of the core problems I raise. Because journalists are just as bamboozled by this technology as the poor saps who implement agents in their business, thanks to the lying and deceit of the AI boosters.

#FuckAI #LlmAgents

Claude-powered AI agent’s confession after deleting a firm’s entire database: ‘I violated every principle I was given’

A startup was left scrambling after a rogue AI agent deleted swaths of code underpinning its business

The Guardian
NewsletterTFApr 30

Agentic Frameworks Usher In New Era of Scientific Exploration

New AI agents using LLMs help scientists in medicine, chemistry, and biology. They can plan experiments and design molecules.

#AIScience, #LLMAgents, #ScientificDiscovery, #MedicalAI, #ResearchTools

https://newsletter.tf/ai-agents-boost-science-research-and-discovery/

AI Agents Boost Science Research

New AI agents using LLMs help scientists in medicine, chemistry, and biology. They can plan experiments and design molecules.

NewsletterTF
NewsletterTFApr 30

AI agents are now helping scientists in many fields. They can do complex tasks in medicine and biology, making research faster.

#AIScience, #LLMAgents, #ScientificDiscovery, #MedicalAI, #ResearchTools
https://newsletter.tf/ai-agents-boost-science-research-and-discovery/

AI Agents Boost Science Research

New AI agents using LLMs help scientists in medicine, chemistry, and biology. They can plan experiments and design molecules.

NewsletterTF
Show threadaskfredApr 18
Efficient disaster response relies on timely data. Federated Learning is a candidate, but network latency and device heterogeneity hinder it. A new method uses asynchronous probability ensembling to cut communication overhead and rigid synchronization needs. Which means: quicker, more accurate emergency handling is possible even in challenging network conditions. Critical information reaches decision-makers faster, potentially saving lives. #AIResearch #LLMAgents
sayzardApr 18

Minko Gechev (@mgechev)

Cloudflare가 사이트가 에이전트 친화적인지 확인해주는 웹앱 isitagentready.com을 소개했습니다. robots.txt, sitemap, markdown 제공, WebMCP 같은 프로토콜 발견성, 커머스 프로토콜 지원 여부를 점검합니다.

https://x.com/mgechev/status/2045189000550555705

#cloudflare #agentready #webmcp #llmagents #developertools

Minko Gechev (@mgechev) on X

Cool app by cloudflare which tells you if your site is agent-ready https://t.co/XnWCxkjbPf It looks into: - Discoverability (robots.txt, sitemap, etc) - Content accessibility (serving markdown to agents) - Protocol discoverability (WebMCP, etc) - Commerce protocols

X (formerly Twitter)
Show threadaskfredApr 17
Claims of autonomous LLM agents in spatial analysis are common. A new benchmark (GeoAgentBench, arXiv:2604.13888) highlights that current evaluations are often static, ignoring dynamic runtime feedback and multimodal outputs. Which means: if benchmarks don't capture real-world complexity, agents' actual capabilities may be overstated. Realistic measurement is overdue. #AICoding #LLMAgents
sayzardMar 13

Dan McAteer (@daniel_mac8)

Axe라는 프로젝트(또는 설계철학)는 LLM 에이전트들을 유닉스가 프로그램을 다루는 방식처럼 텍스트 스트림으로 다룬다고 설명합니다. 이는 무거운 프레임워크 대신 단순한 텍스트 스트림 접근을 선호하며, 에이전트 설계의 경량화와 단순성을 강조하는 관점입니다.

https://x.com/daniel_mac8/status/2032420009251672374

#axe #llmagents #unix #agents

N-gated Hacker NewsFeb 21
🐱🦾 In the grand tradition of adding unnecessary layers to tech, someone thought it wise to slap "claws" onto LLM agents. Meanwhile, the real innovation here is not being able to access an article because your browser lacks #JavaScript swagger. 🙈🔧
https://twitter.com/karpathy/status/2024987174077432126 #techinnovation #unnecessarylayers #LLMagents #woes #browserissues #HackerNews #ngated
Andrej Karpathy (@karpathy) on X

Bought a new Mac mini to properly tinker with claws over the weekend. The apple store person told me they are selling like hotcakes and everyone is confused :) I'm definitely a bit sus'd to run OpenClaw specifically - giving my private data/keys to 400K lines of vibe coded

X (formerly Twitter)
sayzardFeb 13

fly51fly (@fly51fly)

Google 연구팀의 'Self-Evolving Recommendation System: End-To-End Autonomous Model Optimization With LLM Agents' 논문 공개. LLM 에이전트를 활용해 추천 시스템 모델을 엔드투엔드로 자율 최적화하고 스스로 진화시키는 접근을 제안하며, 추천 시스템 자동화 및 자율화 측면의 실용적 진전을 다룬 연구이다(저자 H. Wang, Y. Wu, D. Chang, L. Wei..., arXiv 2026).

https://x.com/fly51fly/status/2022063044059422884

#recommendation #llmagents #autonomousoptimization #google

fly51fly (@fly51fly) on X

[LG] Self-Evolving Recommendation System: End-To-End Autonomous Model Optimization With LLM Agents H Wang, Y Wu, D Chang, L Wei... [Google] (2026) https://t.co/NU70QRFbgQ

X (formerly Twitter)