The resource shelf

The official spec, SDKs, and quickstarts for the agent↔tools standard. Start with the server quickstart.

Linux Foundation-governed spec (v1.0) for cross-vendor agent delegation — task lifecycle, discovery, agent cards.

Event-based protocol for streaming agent state into frontends: thinking steps, tool results, interactive components.

The vendor-neutral telemetry schema for model calls and agent spans. Still maturing — pin the version you adopt.

The contract language behind every tool definition and structured output you will ever write.

API reference, tool use, and the agent SDK. The MCP ecosystem's home turf.

API + Agents SDK docs; the OpenAI-compatible protocol is the de-facto local-serving interface too.

Gemini API docs and agent tooling.

Open-weights flagship family — downloads, licenses, and docs.

Strong open-weights and hosted models out of Europe; good tool-calling small models.

Alibaba's open-weights family — consistently strong at agentic/tool-use benchmarks per size class.

CPU/GPU inference anywhere, first-class ARM64; llama-server speaks OpenAI-compatible.

The GPU-throughput serving standard for open-weights in production.

The developer-convenience layer for local models — one command to run.

Open-source LLM gateway: one API for 100+ providers, keys, budgets, fallbacks.

Open-source LLM tracing, evals, and prompt management; self-hostable.

Open-source tracing + evaluation, OTel-native.

Tracing and evaluation from the Weights & Biases stable.

Config-driven eval runner — a good step up when your hand-rolled harness grows.

Home of the LLM Top 10 and the Top 10 for Agentic Applications (2026) — the peer-reviewed agent threat taxonomy. Map your threat model to it.

The US reference framework for AI risk — voluntary, widely cited in enterprise governance programs.

Regulation (EU) 2024/1689. GPAI obligations phased in from Aug 2025; high-risk obligations from Aug 2026.

The US banking model-risk letter whose discipline underpins our governance framework — public document.

The paper behind the think→act→observe loop every agent runs.

Models teaching themselves when to call tools — tool use becomes mainstream.

Agents improving via verbal self-feedback — the seed of self-correction loops.

The Smallville simulation — memory, reflection, and planning working together.

The benchmark that made coding agents measurable on real GitHub issues.