Dependabot can now keep AI agent skills and prompts up to date

GitHub's Dependabot gained Agent Package Manager (APM) support on 3 November 2026, treating AI agent skills, prompts and instructions as dependencies that need updating like any other software.

GitHub announced on 3 November 2026 that Dependabot version updates now support Agent Package Manager (APM). GitHub describes the feature as a way to keep AI agent skills, prompts and instructions current.

What the changelog says

The interpretation

Until recently, the instructions and skills you gave an AI coding agent were files someone wrote once and forgot. Treating them as versioned dependencies is a quiet but important shift: they now have owners, versions and update pull requests, just like libraries do.

That matters because those files steer what the AI builds in your product. A stale or silently changed skill file can change code quality without anyone touching the application. The flip side is that automatic update pull requests are only as safe as the person who reviews them. The changelog is explicit that no security-update coverage comes with this.

So what

If your supplier uses AI agents on your codebase, ask where their agent instructions live, whether they are versioned, and who approves changes to them. It is a cheap question and it separates disciplined AI-assisted development from improvisation. BuildApps keeps AI workflows documented and reviewed. See AI-assisted development, custom software development, or start a project.

← All research

More research.

GitHub now reports how often developers act on Copilot code review comments

On 5 November 2026 GitHub added code review metrics, including an addressed rate, to the Copilot usage metrics API. For AI software development buyers, it is the first first-party number for how much AI review feedback is acted on.

Read ›

GitHub Copilot is now inside Xcode's own coding assistant: what iOS app owners should ask

GitHub put Copilot into Xcode's built-in coding assistant on 4 November 2026, in public preview with Xcode 27.3 beta. AI-assisted iOS app development is now a built-in choice of agent, not a plug-in.

Read ›

GitHub's Copilot usage API now separates bring-your-own-key and custom models: AI spend visibility for software buyers

GitHub's Copilot usage metrics API now reports bring-your-own-key (BYOK) and enterprise custom model usage separately, as of 3 November 2026. Teams using AI coding tools can see more of where usage goes.

Read ›

GitHub Copilot arrives in JetBrains Air: parallel AI coding sessions, cloud tasks billed to your Copilot plan

GitHub released Copilot inside JetBrains Air on 3 November 2026, letting teams run several AI coding agent sessions in parallel, locally or in cloud environments. AI-assisted development is becoming a choice of agent and workspace, not one tool.

Read ›

Claude Code 2.1.318: AWS cross-account cost attribution, no more automatic IDE extension installs

Anthropic's 4 November 2026 Claude Code release adds per-developer AWS cost attribution through the Claude apps gateway and stops auto-installing the IDE extension from IDE terminals. AI coding tool cost control is becoming a product feature.

Read ›

GitHub will block pull_request_target by default on 2 December, with test blackouts on 9 and 16 November

GitHub is switching off the pull_request_target workflow trigger by default in public repositories from 2 December 2026. Software teams that automate with GitHub Actions, including AI coding agents, need to check their pipelines now.

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