diff --git a/CLAUDE.md b/CLAUDE.md
index 53e7bcd..58631a8 100644
--- a/CLAUDE.md
+++ b/CLAUDE.md
@@ -19,8 +19,8 @@ docs it indexes.
   commit. Every commit deployable. No feature branches or PRs for solo/agent work — the only
   exception is a short-lived `agent/<desc>` branch when another agent is *simultaneously*
   active on this repo, merged within the same session.
-- **Run the quality gate before every push.** `task check` (once a Taskfile exists). CI is the
-  gate, not branch protection — do not enable branch protection on this repo.
+- **Run the quality gate before every push.** `task check`. CI is the gate, not branch
+  protection — do not enable branch protection on this repo.
 - **Conventional commits.** `feat:`, `fix:`, `chore:`, `docs:`, `refactor:`. Subject line says
   what; body says why.
 - **Language is Go (ADR-001).** Do not introduce Python or a second language. If you think you
@@ -63,6 +63,40 @@ These caused real mistakes that were caught and corrected; the corrections are l
 - **BDD.** The `docs/use-cases/*.feature` files are the behavior spec. New behavior gets a
   scenario; the use-case core is tested through fake adapters, not live YouTube/brain.
 
+## Skills (engineering discipline)
+
+Skills live in the canonical library `mathias/skills` and are wired into this repo as
+**gitignored symlinks** — do not commit them, do not add a skills manifest to this repo.
+
+- **Install/refresh:** `task skills` (or `curl -fsSL https://gitea.d-ma.be/mathias/skills/raw/branch/main/install.sh | bash`). Idempotent.
+- **Most relevant skills for Tapir work:**
+  - `tdd`, `atdd`, `test-design` — the build is test-first; the `.feature` files are the ATDD spec.
+  - `clean-code`, `solid` — the ports-and-adapters structure depends on these.
+  - `gitea-ci` — for the CI workflow (and its act_runner gotchas) and the TBD section.
+  - `debug` — when `task check` goes red for a non-obvious reason.
+  - Reviewer-side: `code-review`, `refactoring`, `cognitive-load`.
+  - Session close: `session-retrospective` (surface learnings into brain before context is lost).
+
+## Current build state (start here for the first task)
+
+The repo is **scaffolded and intentionally RED**:
+
+- Clean Architecture skeleton exists: `internal/domain` (entities), `internal/ports`
+  (interfaces), `internal/usecase` (engine), `cmd/tapir` (entrypoint stub),
+  `internal/adapters` (empty — concrete adapters go here).
+- `usecase.Engine.ProcessNewVideo` returns `ErrNotImplemented`.
+- `test/acceptance/summarize_new_video_test.go` translates the first two Gherkin scenarios and
+  **fails** against the stub. `task check` is therefore red on `test`.
+- **First build task:** implement `ProcessNewVideo` (resolve transcript -> summarize -> deliver to
+  sinks | skip on no-transcript) to make the acceptance tests green, following the `.feature`
+  files. Then add the AI-router `Summarizer` (copy `llm` per ADR-004), the YouTube `VideoSource`
+  adapter (captions-first), and the store + brain sinks.
+
+**Unverified setup items** (see `docs/homelab-integration.md`, marked `confirm`): the Go version
+in `go.mod` (1.23 — match the koala runner; estate elsewhere uses 1.26.1), the brain-mcp URL, the
+exact ESO secret-ref naming, and the summarization model alias. Resolve against the live cluster
+before depending on them, and pin answers back into `docs/homelab-integration.md`.
+
 ## Provenance (where this design came from)
 
 - The reuse decisions came from **Spike S5**, recorded at