tooling

SpadeBox is an open-source sandboxed tools and JavaScript runtime for AI agents, with Rust core plus JavaScript and Python bindings. It provides file, network, and code-execution tools, supports MCP, and includes examples for JavaScript, Rust, and Python.

Intencion is a product analytics and observability platform for AI agents. It tracks user intent, tool calls, run outcomes, and resolution trends, then turns failures or unmet requests into Linear or GitHub issues.

Mirrors is a staging and replay environment for AI agents. It rebuilds the systems an agent calls, replays real sessions against prompts, tools, and model changes, and surfaces regressions before deployment.

Recurse is a serverless harness for building specialist agents and deploying them as tools, MCPs, or bots. The page shows a concrete manifest-based workflow, validation of inputs and outputs, and examples for tasks like level design and RNA sequence design.

A long-form technical article on building ship-ready AI agents in Rails with the Anthropic Ruby SDK, covering tool design, agent loops, MCP servers, human approval, testing, and cost control.

Autonomy is a self-directed AI agent framework under development by Bill Liu. The page describes a turn-based AgentLoop, managed action execution, beam-search candidate ranking, procedure skills, post-run learning, and event-sourced run history, with support for multiple AI providers and extensible toolsets.

Go packages for building agents in Go, with a streaming runner, tool dispatch, guardrails, compaction, provider adapters, and verified completion helpers.

Myli is a provider-neutral Python harness for multi-agent visual design workflows. It defines typed design specs, validates JSON candidates at runtime, supports tools and middleware, and can render candidates for visual review with event and trace reporting.

Personal site for an engineer writing about AI and dev tooling, with visible open-source projects for agent orchestration and data pipelines. The page highlights LLAssembly, described as code-driven orchestration for sub-agents and tool-call workflows, plus other related projects and writing.

Python toolkit for building LLM agents with MCP servers, skills folders, memory, streaming, and metrics support.

A dated research report from Vouch that analyzes the MCP registry at scale, including corpus growth, transport split, credential classes, schema quality, and unreachable servers. It also points to queryable live data and the methodology behind the counts.

A blog post describing an evaluation harness for comparing agentic coding tools and prompts across realistic SWE tasks, with token-cost results and model-specific behavior notes.

A research-style blog post comparing grep with LSP-backed navigation in coding-agent workflows, showing how tool interface shape affects agent behavior and task success.