Reading
Agent Foundry is an open-source AI Agent plugin registry and CLI tool that supports installing unified plugins across multiple AI coding assistant platforms including Codex, Claude Code, Cursor, and Copilot, solving the configuration management challenges caused by the fragmentation of AI tools.
Section 01
Agent Foundry is an open-source AI Agent plugin registry and CLI tool that supports installing unified plugins across multiple AI coding assistant platforms (Codex, Claude Code, Cursor, Copilot etc.). It addresses the configuration management challenges caused by the fragmentation of AI tool ecosystems, allowing developers to write plugins once and use them across platforms.
Section 02
Mainstream AI coding assistants have independent extension mechanisms:
.codex/plugins/ directory and marketplace.json..claude/ directory and supports MCP (Model Context Protocol) servers.Section 03
Agent Foundry's core design includes:
plugins/ directory with plugin.yaml (metadata), skills/ (reusable skill definitions), agents/ (complete agent definitions), and mcp/ (MCP server configs).registry/plugins.yaml as a central index for plugin discovery, version management, and dependency resolution.Section 04
Install CLI: Use pipx install git+https://github.com/josemiguelmelo/agent-foundry.git (recommended) or source installation.
Install Plugins: Global (e.g., agent-foundry install codex code-reviewer), project-level (add --in-project), or local repo (add --repo path).
Manage Plugins: Uninstall (e.g., agent-foundry uninstall codex code-reviewer), validate registry (agent-foundry validate-plugins), or check help (agent-foundry --help).
Develop Plugins: Create skeleton (agent-foundry create-plugin my-awesome-plugin), remove plugin (agent-foundry remove-plugin my-awesome-plugin).
Section 05
Skill Design Principles: Clear trigger conditions, focused职责, configurable parameters, and complete examples. Agent Orchestration Patterns: Pipeline (doc→code→test→review), decision tree (based on complexity), multi-expert会诊 (parallel reviews by specialized agents). MCP Integration: Support tool discovery, state management, and security boundaries as per Anthropic's Model Context Protocol standard.
Section 06
Section 07
Current Limitations: Platform ability differences, Claude Code's lack of project-level installation, Copilot's closed ecosystem restrictions, and version compatibility issues. Future Plans: Support more tools (Continue.dev, Aider, Supermaven), build a community plugin market, provide a GUI editor, add automation testing, and enable cloud sync for plugin configurations.
Section 08
| Feature | Agent Foundry | Promptfoo | LangChain CLI |
|---|---|---|---|
| Goal | Multi-platform plugin management | Prompt testing & evaluation | Application development framework |
| Supported Platforms | Codex/Claude/Cursor/Copilot | General | Custom applications |
| Plugin Format | Standard YAML | Any text | Python code |
| Ease of Use | High (CLI-driven) | Medium | Low (requires programming) |
| Agent Foundry's unique value: It bridges existing AI tools instead of replacing them, solving the 'last mile' interoperability problem. |
Keep going with more reads from the same topic.
Nornir MCP Server is an enterprise-level server based on the Model Context Protocol (MCP). It seamlessly integrates large language models (such as Claude) with the Nornir network automation framework, supporting natural language orchestration for multi-vendor network devices (Cisco, Arista, Juniper, etc.), and providing production-grade features like a dual-engine architecture (NAPALM + Netmiko), intelligent filtering, and a secure sandbox.
Bibliothèque Française LLM is a structured indexing and annotation project for French public domain literature designed specifically for large language models (LLMs). It integrates multiple authoritative sources such as DraCor, Common Corpus, and Wikisource, providing metadata indexing categorized by genre, author, and era, as well as in-depth annotations for dramatic texts (including characters, lines, stage directions, etc.). Its aim is to enable LLMs to efficiently read and understand classic French literary works.
Splinter is a minimalist, high-performance key-value (KV) and vector storage system enabling zero-latency inter-process communication via shared memory and atomic operations. With only 766 lines of core code, it supports millions of operations per second and 768-dimensional vector storage, offering a new architectural approach for local LLM inference and data-intensive applications.
libmlxforge is an embeddable MLX large language model (LLM) inference engine designed specifically for Apple Silicon. It provides a unified C ABI interface, supports calls from Node.js, Swift, and Rust, and features continuous batching, streaming output, JSON-constrained structured output, and embedding vector generation.