Skip to content

CLAUDE.md

Latest commit

 

History

History
52 lines (39 loc) · 2.46 KB

CLAUDE.md

File metadata and controls

52 lines (39 loc) · 2.46 KB
Raw

CLAUDE.md

This file provides Claude Code (claude.ai/code) with context for working in this repository.

For the comprehensive convention guide that applies to all AI agents (Claude, Cursor, Aider, Copilot), see AGENTS.md. This file captures only Claude-specific guidance.

Quick orientation

This is macadmin-toolkit — a curated personal collection of scripts, Claude skills, extension attributes, and Terraform patterns for managing macOS fleets via Jamf Pro. Most of it started life inside a private MDM repo at a previous role and has been re-implemented or generalized here.

The repo mixes several languages — Python (under packages/), bash (under components/scripts/shell/), HCL/Terraform (terraform/), and PowerShell. Each has its own conventions; see AGENTS.md for details.

When asked to add a new component

  1. Identify the right home. Scripts → components/scripts/<lang>/. Extension attributes → components/xattrs/. Terraform-templated content → components/configs/<area>/.
  2. Check components/vendor-provided/ before writing something from scratch — there may already be a third-party version.
  3. If the component is delivered via Jamf, write the corresponding Terraform resource in terraform/<domain>.tf and use templatefile() if any content needs to be baked in at deploy time.
  4. Follow the existing fake-data convention. Sample data uses Marvel / Avengers themes (Stark Industries, Wakanda Embassy, etc.) — keep it consistent so the repo reads as a single coherent example rather than a mishmash of placeholders.

When asked to add a Claude skill

Skills live under .claude/skills/<skill-name>/. Each skill is a directory containing at least a SKILL.md describing what it does and when it should be invoked. Refer to existing skills as templates before adding new ones.

Don't

  • Don't commit real Jamf URLs, API credentials, or any organizationally-identifying data. Use Avengers fake data.
  • Don't modify files under docs/ — those are read-only git submodules of upstream wikis.
  • Don't add new Python top-level dependencies without considering whether they fit in packages/common/ (shared library) vs packages/jamfy/ (CLI-specific).

Useful commands

make help          # list every Make target
make format        # auto-format Python + Terraform
make lint          # check formatting and run ruff
make test          # run pytest
make docs-update   # refresh submoduled docs/