54-tool 'Claude Code' resource surfaced

A GitHub gist called the “Claude Code Resource Bible” appeared this week with 54 links packed into one page, and the list is organized like a starter kit instead of a reading list. The first section points to Anthropic’s own docs and partner network, then it jumps straight into server repos, skills directories, browser tools, and terminal multiplexers. (gist.github.com) That matters because Claude Code is not just a chatbot that suggests code in a text box. Anthropic describes it as an “agentic coding tool” that can read a codebase, edit files, run commands, and integrate with development tools from the terminal, desktop app, browser, and integrated development environments. (code.claude.com) The word “agentic” is the key to why lists like this keep spreading. In Anthropic’s setup, the model is allowed to take actions like opening files, running tests, or using command-line tools, so the useful part is no longer just the model itself but the collection of tools around it. (code.claude.com) Anthropic now exposes that same pattern as a software kit called the Claude Agent Software Development Kit. Its docs say developers can build agents in Python or TypeScript that read files, run commands, search the web, and edit code using the same tool loop that powers Claude Code. (platform.claude.com) Once that exists, a “resource bible” becomes a map of attachments for the agent. In the gist, one cluster is Model Context Protocol servers, which are connectors that let an agent reach outside its own chat window into places like GitHub, Slack, Notion, PostgreSQL, Supabase, Figma, and Sentry. (gist.github.com) Another cluster is about giving the agent eyes and hands in a browser. The list includes Playwright for browser automation and Firecrawl for web scraping and search, which turns “go check this page and pull the data” from a human task into a tool call. (gist.github.com) A third cluster is about running more than one worker at a time. The gist includes cmux, gmux, claude-squad, and dmux, all of which are built around the same idea: split work across parallel agent sessions the way a manager splits work across interns sitting at different desks. (gist.github.com) Anthropic’s own product page shows why that idea is catching on. The company says Claude Code can handle multi-file changes, run tests, monitor continuous integration failures, and even work across tools like the GitHub command-line interface, while teams shift toward “continuous orchestration” of multiple agents in parallel. (anthropic.com) The audience for a list like this is wider than software engineers. Anthropic’s product page now explicitly pitches Claude Code to founders, product managers, designers, and operations teams, and says non-engineering users are already building prototypes, internal tools, and personal projects in plain language. (anthropic.com) So the news is not that one person posted 54 links on GitHub. The news is that a one-page kit for agents now looks a lot like a modern software stack: one model, a pile of connectors, a browser robot, a database bridge, and a few ways to run several jobs at once. (gist.github.com)