Write once, reuse across tools

Look again at where you were told to put the skill: .github/skills, .claude/skills, or .agents/skills. Those locations reflect a shared, increasingly adopted Agent Skills format. The SKILL.md you just wrote is designed to be portable, but support still depends on the tool, version, location, and feature surface.

One file, supported surfaces

The audit-event skill can be reused across Copilot surfaces that support Agent Skills:

• VS Code - where you’ve worked this whole chapter, Copilot pulls it in when a task matches the description.
• The Copilot CLI - current versions support the shared skill locations; verify the version and discovery output.
• The cloud coding agent - repository skills may be available to the agent, subject to the cloud surface and repository policy.

So the procedure you wrote once travels with the work, whoever - or whatever - picks it up.

The convergence payoff

Now the part that reaches past Copilot entirely. Back in the Rules chapter you saw the first convergence beat: the AGENTS.md and CLAUDE.md files you’d written for another agent were read natively by Copilot, no translation. The same context, honored by a different tool.

Skills are the second beat of the same story. The SKILL.md you wrote for the audit-event convention here follows the same format used in the Claude Code course and the Codex course. A skill authored there may work here unchanged, but verify each tool’s supported locations and version before treating that as a guarantee.

That’s the quiet thesis of this whole site arriving twice over. The hard part of working with agents isn’t the tool - it’s closing the gap between what the agent knows and what your team knows: your conventions, your procedures, the audit-event shape everyone on the Platform squad has in their head. You close that gap by writing it down once. Rules proved your always-on context survives the jump between tools. Skills prove your invokable procedures do too. You’re not re-teaching the audit convention to each new agent you adopt - you taught it once, in a standard format, and every conforming tool already speaks it.

Where this leaves the work

You’ve now packaged the team’s audit-event convention as a portable skill Copilot invokes on its own, in both repos, in a format your other agents already understand. The procedure is handled. What’s still missing is reach: the skill knows how to record an event, but it can’t yet see the internal systems the work depends on - the issue tracker, the service registry that lists which of the dozen consumers a shared-lib change touches. To give Copilot a live line into a system outside the repo, you wire up an MCP server. Next chapter: extending Copilot over MCP.