Claude Code Mods: the AI coding tool can now be rewritten from the inside

Anthropic has added Mods to Claude Code, a plugin system that lets developers change how the AI coding tool behaves. Mods are not sandboxed, so what your supplier installs matters.

Claude Code, one of the most widely used AI coding tools, gained a built-in plugin system called Mods in version 2.1.287, released on 1 October. Coverage from The Decoder on 3 October describes Mods as JavaScript or TypeScript functions that hook into what the tool does: tool calls, prompts, even what appears on screen. Anthropic says some of its own features, such as the /diff command, are already built this way.

What is actually new

Until now, developers could extend Claude Code through hooks and external connectors that ran outside the tool. Mods run inside it. They can add panels, intercept actions before they happen, and add new commands. They work in the command line and desktop app, and partly in the VS Code extension. The first official example, “You Should Know”, runs a second agent that watches the first and flags things the developer may have missed.

The part buyers should notice

Mods run with the developer’s own permissions and are not sandboxed. Anthropic’s own advice is to install them only from trusted sources. Organisations can control which Mods are allowed to load, but only if someone has set that up.

That puts a familiar problem in a new place. A plugin on a developer’s laptop can see what the developer sees: your code, your credentials, your test data. The risk is not that Mods are bad. It is that AI tooling is now an ecosystem of third-party add-ons, with the same supply-chain questions as any other software.

What to ask a supplier

So what

Customisable AI tooling is good news for teams that use it well. It lets a disciplined studio build review steps and guardrails into the way agents work. It is bad news for teams that treat “we use Claude Code” as a complete answer. The tool matters less than the process around it. See how we run AI-assisted development with experienced engineers accountable for every change, or start a project and ask us how we govern our tooling.

← All research

More research.

GitHub's Copilot metrics now count OpenCode and Claude Code traffic: who really uses which AI coding tool

On 23 October 2026 GitHub added a per-tool breakdown to the Copilot usage metrics API, so tools such as OpenCode and Claude Code are counted even without editor telemetry. It shows how mixed the AI coding tool stack inside software teams has become.

Read ›

GitHub Copilot retires Claude Opus 5 and GPT-5.6 Terra on 16 November: the model-churn bill for buyers

GitHub has announced that Claude Opus 4.8, Claude Opus 5 and GPT-5.6 Terra leave every Copilot experience on 16 November 2026. For anyone commissioning AI-assisted development, model retirement is now a monthly event to plan around.

Read ›

GitHub Copilot now benchmarks how 'agent-first' your developers are against your industry

On 22 October 2026 GitHub added opt-in industry comparisons to its Copilot impact dashboard, ranking firms by share of agent-first developers and by the merge rate of agentic pull requests. Here is how to read the numbers when you are buying software.

Read ›

Apple's new bait-and-switch rule 2.5.19: what it means for AI-built and fast-shipped apps

On 16 October 2026 Apple updated its App Review Guidelines: apps that change after approval can be removed from the App Store and the Developer Program, and AI-generated intimate imagery is now explicitly covered. Here is what founders commissioning an iOS app should check.

Read ›

GitHub Copilot agentic autofix is generally available: AI now fixes security alerts, and bills for it

GitHub made agentic autofix for code scanning alerts generally available on 15 October 2026. Copilot can now fix up to 25 security alerts in one pull request, using AI credits your supplier or your team pays for.

Read ›

GitHub AI Scan for pull requests goes GA: AI security checks for PHP, Terraform and Dockerfiles

GitHub's AI Scan for pull requests became generally available on 15 October 2026, checking languages the standard scanner misses. Findings are advisory and cannot block a merge, which matters when you buy software.

Read ›

Start a project

Building with AI? Let's talk.

Tell us what you are building. We'll give you an honest view of the right approach.

Start a projectSee our work