← ClaudeAtlas

adopt-configlisted

Discover pre-existing Claude config on this machine (files install.sh/install.py won't overwrite, plus other known sources) and integrate the worthwhile parts into this repo. Two-stage flow (adopt inbox -> adopted archive) with a tracked ledger so nothing is evaluated twice. Use when setting up on a new machine, when the installer reports "exists and is not a symlink", or to learn from prior config/memory.
noctua84/nescio-ai · ★ 0 · AI & Automation · score 73
Install: claude install-skill noctua84/nescio-ai
# /adopt-config Adopt Claude configuration that already exists on this machine into the repo, so the setup can absorb — and learn from — what was here before it was managed. ## The two-stage model | Location | Role | Tracked? | |----------|------|----------| | `eval/adopt/<ts>/` | **Inbox** — discovered, needs evaluation | no (gitignored) | | `eval/adopted/<ts>/` | **Archive** — processed runs | no (gitignored) | | `memory/adoption-log.md` | **Ledger** — what was adopted, when, and its disposition (≤150 lines) | **yes** | The ledger is the dedup source: it's keyed by a short content hash, so a scan on any machine skips sources already recorded terminal (`integrated` / `no-change` / `dropped`). `pending` items re-surface into the inbox until resolved. The raw `eval/` trees stay machine-local — only the compact ledger syncs. ## When to use - The installer printed `! … exists and is not a symlink — back it up and rerun, skipping.` for `CLAUDE.md`, `settings.json`, or `settings.local.json`. - Setting up on a machine that already had Claude Code / Desktop use. - Pulling prior per-project memory (`~/.claude/projects/*/memory/`) into the repo. ## Step 1 — Discover & stage (mechanical) ```bash python scripts/adopt_existing_config.py # or python3 ``` Copies not-yet-processed sources into `eval/adopt/<ts>/` with a `MANIFEST.md` (and a machine-readable `sources.json`). Skips sources already recorded terminal in the ledger, and sources already symlinked into the repo. Prin