plan: initialization.md — §0 v2 deprecation banner + §11 verification block

# {tasking-branch} — Initialization agent checklist

**Audience:** the **Initialization** agent (§9.4 of the meta-plan),

run once before the three-agent orchestration starts; also re-runnable

whenever the environment changes (new machine, rotated PAT, new

remote).

## 0. v2 deprecation banner

> **In v2 the standalone Initialization agent is deprecated.** Its

> work is now performed by **phase 1 of the Administrator agent**;

> see [Administrator-prompt.md](Administrator-prompt.md) and

> `plan/orchestration.md` §6.4 / §9.6 / §13.1.

>

> This file remains **canonical for the checklist content** —

> Administrator phase 1 walks §1-§10 below, plus the §11

> Target-branch dependency verification block (added in v2 to catch

> the `pytest-timeout` gap that leaked into the dry-run delta cycle —

> see `dry-run-Analyst-findings.md` F1). Re-init flows (credentials

> rotated, machine moved, `{base-branch}` switched) re-walk this

> document the same way.

>

> The v1 standalone Initialization session remains an acceptable

> fallback — see [Initialization-prompt.md](Initialization-prompt.md).

> New deployments should start the Administrator agent first per

> `plan/orchestration.md` §13.1 instead.

---

**Audience:** **Administrator phase 1** (`plan/Administrator-prompt.md`),

run first on each machine before the three worker agents start; or

the v1 standalone Initialization agent (deprecated, see §0). Also

re-runnable whenever the environment changes (new machine, rotated

PAT, new remote).

**Purpose:** verify that every local feature, credential path, hook,

and remote operation the orchestration depends on actually works

- Switched `{base-branch}` to a branch with different protection

  rules.

- Upgraded `git`, `pixi`, `gh`, or `glab` to a new major version.

- Per-cycle plugins (e.g., `pytest-timeout`) were added or removed.

The checklist is idempotent: re-running on a clean environment

produces the same report and cleans up the same way.

---

## 11. Target-branch dependency verification (v2; per redesign §15.2)

Run this block **after** §8 ("Test environment") and **before**

declaring "Phase 1 complete." (or, in the v1 standalone path,

"Ready to launch orchestration.") in §9. It catches the F1 finding

(missing `pytest-timeout` plugin) at protocol-rehearsal time

instead of leaking it into the Integrator's `dry-run-delta` cycle.

1. **Confirm the active pixi env corresponds to `{base-branch}`**

   (`new_workflow_ui_plan` for this effort):

   ```bash

   cd lr_reduction

   git rev-parse --abbrev-ref HEAD          # expect {base-branch}

   pixi info | grep -E '^(Project|Environments)'

   ```

2. **Probe each test-runtime plugin the protocol requires.** For

   the dry-run delta path and any production cycle that uses

   `--timeout`:

   ```bash

   pixi run python -c "import pytest_timeout; print('pytest-timeout', pytest_timeout.__version__)" \

     || echo "MISSING: pytest-timeout — needed for dry-run-delta. \

              See plan/orchestration-v2-redesign.md §15.1 for install."

   ```

3. **Confirm the canonical test commands accept the full argument

   set the Integrator will use.** `--collect-only` alone (per §8)

   verifies pytest imports; `--timeout=1` transitively verifies

   `pytest-timeout` is loaded:

   ```bash

   pixi run test-reduction -- --collect-only --timeout=1 -k 'test_does_not_exist' \

     2>&1 | tail -5

   ```

   Expect a clean "0 tests collected" line — not an arg-parse error

   (exit 4) and not an environment failure.

4. **Dry-run only**, when `{integrator-test-cmd} = test-dry-run`:

   confirm the dry-run pixi task exists:

   ```bash

   pixi run test-dry-run -- --collect-only 2>&1 | tail -5

   ```

   Expect "0 tests collected" cleanly (the synthesized test files

   don't exist yet on a fresh `{base-branch}`; the *task itself*

   must exist and be invokable). If the task is missing, that's a

   `pyproject.toml` gap — see `plan/orchestration-v2-redesign.md`

   §14.

5. **Read `pyproject.toml`'s `[tool.ruff.lint]` ignore list** and

   confirm it is non-empty (per `dry-run-Initialization-findings.md`

   §2; missing ignore list ⇒ 780 ruff violations on commit, breaking

   pre-commit hooks):

   ```bash

   grep -A30 '\[tool\.ruff\.lint\]' lr_reduction/pyproject.toml \

     | grep -E '^\s*(ignore|extend-ignore)' \

     || echo "WARNING: [tool.ruff.lint].ignore is empty — pre-commit \

              hooks will reject commits with 780+ violations."

   ```

If any check fails, STOP and hand the user a specific action item.

Do **not** enter Administrator phase 2 (or, for the v1 path, launch

the worker agents) until every check passes.