atlas

1# atlas
2 
3This skill turns a stated need into a **real systems atlas** by SCANNING your
4actual sources — Azure subscriptions (via read-only `az`), git repos, and local
5directories — and process/data mining them, THEN enriching the result against the
6Atlas knowledge graph. It is the brain of the `atlas` plugin. The scan is
7PRIMARY; the graph is SECONDARY. For non-trivial runs it delegates orchestration
8to `babysitter:babysit` using an atlas-specific `.a5c` process; for simple
9lookups it queries the graph directly.
10 
11## 1. Scan-first, graph-second
12 
13The output you want is an evidence-backed inventory of **your** systems — e.g.
14`azure-inventory.json` (every real resource id + RG from `az`),
15`workspace-inventory.json` (real repo/dir scan), `processes.json` (real mined
16CI/CD/IaC/.a5c processes), and a cross-linked `SYSTEMS-ATLAS.md`. Every item must
17cite its REAL source. Generic catalog nodes are NOT the deliverable.
18 
19- **Primary — scan the user's real sources.** Use `Bash` to run READ-ONLY scans:
20  `az` (account/group/resource list + per-service list/show) for Azure;
21  `git` + filesystem (`Read`/`Glob`) for repos and directories. NEVER invent
22  resource ids, regions, SKUs, or file paths — if you didn't observe it in real
23  output, it does not go in the atlas. Only scan the sources named in the need
24  (scoping, not a fallback).
25- **Secondary — the Atlas knowledge graph.** Atlas is a knowledge graph of
26  agents, processes, data models, capabilities, workflows, and wiki pages reached
27…