anthropics/config-guide
Reference guide for configuring @Claude agents — agents, agent scopes, identity profiles, presets, connections, rules, GitHub repositories, and custom instructions. Explains the inheritance model and configuration best practices.
npx skills add https://github.com/anthropics/claude-tag-plugins --skill config-guide@Claude is configured through a small set of layered objects in the claude.ai
admin settings: agents, agent scopes, identity profiles, and the
presets / connections / repos / instructions attached to them. The
layering decides what any given session can see and do.
This guide is currently written for the Slack surface.
This skill is an index. Read the relevant reference file below for the
topic the user is asking about — each is short and self-contained.
| Topic | Reference file |
|---|---|
| Agents, agent scopes, the resolution chain, and per-field inheritance | agents-and-scopes.md |
| Identity profiles, scopes, rules, credentials, repo permissions | identity-profiles.md |
| Presets, connections (OAuth / MCP), how to install one | connections-and-presets.md |
| GitHub repositories and custom instructions | github-and-instructions.md |
| Best practices: profile layout, read-only vs read/write, rollout | best-practices.md |
All reference files live alongside this SKILL.md in the same directory.
After explaining configuration, you can suggest the user run the
debug-plugins skill in a new Slack thread to confirm the change actually
took effect inside the container.
Take anthropics/config-guide from the repository into ~/.claude/skills for personal
use, or into .claude/skills inside a project.
The agent identifies a skill by the name field in its header. Two skills with the
same name cannot sit side by side — one of them will be ignored.