diff --git a/.claude/skills/recipe-upgrade/SKILL.md b/.claude/skills/recipe-upgrade/SKILL.md
index 4cb8861..072a678 100644
--- a/.claude/skills/recipe-upgrade/SKILL.md
+++ b/.claude/skills/recipe-upgrade/SKILL.md
@@ -29,15 +29,25 @@ test** to make a PR green.
 
 ## Procedure
 
+> ### ⚠️ Run abra over a pseudo-TTY (or it FATAs `inappropriate ioctl for device`)
+> `abra` needs a TTY. Over plain `ssh cc-ci 'abra …'` there is none, so many abra commands die with
+> **`FATA inappropriate ioctl for device`** / `the input device is not a TTY`. This is how the working
+> harness avoids it (`runner/harness/abra.py`): wrap every abra call in a pseudo-TTY via util-linux
+> `script`, and pass `-n` (`--no-input`) so it never prompts. **Always invoke abra like this:**
+> ```
+> ssh cc-ci 'script -qec "abra <args> -n" /dev/null'
+> ```
+> (`git`/other commands do NOT need the wrapper — only `abra`.) Do this for every abra command below.
+
 ### 1. Plan the upgrade (research — follow recipe-upgrade-plan methodology)
-On cc-ci, refresh the recipe and see what's available:
+On cc-ci, refresh the recipe and see what's available (each abra call wrapped per the box above):
 ```
 ssh cc-ci 'export PATH=/run/current-system/sw/bin:$PATH; \
   git -C ~/.abra/recipes/<recipe> status --short; \
-  abra recipe fetch <recipe> --force; \
+  script -qec "abra recipe fetch <recipe> --force -n" /dev/null; \
   git -C ~/.abra/recipes/<recipe> fetch origin main; \
-  abra recipe versions <recipe> -m; \
-  abra recipe upgrade <recipe> -m -n'
+  script -qec "abra recipe versions <recipe> -m -n" /dev/null; \
+  script -qec "abra recipe upgrade <recipe> -m -n" /dev/null'
 ```
 - **Dirty worktree on cc-ci → abort this recipe** (status `SKIPPED — dirty-worktree`); don't prompt.
 - **Reconcile the mirror first (always, even if up to date).** Run:
@@ -61,11 +71,13 @@ ssh cc-ci 'export PATH=/run/current-system/sw/bin:$PATH; \
   config changes, risks. (No human review gate — proceed straight to implement.)
 
 ### 2. Implement the upgrade (follow recipe-upgrade-apply methodology, on cc-ci)
-On cc-ci's `~/.abra/recipes/<recipe>`:
-- `abra recipe upgrade <recipe> -n` for the tags it can bump; hand-edit `compose.yml` for any tags it
-  won't, plus the config/env/volume/label changes the plan calls for.
+On cc-ci's `~/.abra/recipes/<recipe>` (wrap every abra call per the pseudo-TTY box above —
+`ssh cc-ci 'script -qec "abra … -n" /dev/null'`):
+- `script -qec "abra recipe upgrade <recipe> -n" /dev/null` for the tags it can bump; hand-edit
+  `compose.yml` for any tags it won't, plus the config/env/volume/label changes the plan calls for.
 - Bump the `coop-cloud.${STACK_NAME}.version` label to the new version string from the plan.
-- `abra recipe lint <recipe> -C` — fix obvious lint errors; if unfixable, record it and continue.
+- `script -qec "abra recipe lint <recipe> -C -n" /dev/null` — fix obvious lint errors; if unfixable,
+  record it and continue.
 - Commit on a branch: `git commit -m "chore: upgrade to <new-version>"` (the commit message drives the
   PR branch name `upgrade-<new-version>`). Do **not** push to upstream; do **not** tag-push.
 
diff --git a/.claude/skills/upgrade-all/SKILL.md b/.claude/skills/upgrade-all/SKILL.md
index f35428c..0393256 100644
--- a/.claude/skills/upgrade-all/SKILL.md
+++ b/.claude/skills/upgrade-all/SKILL.md
@@ -30,12 +30,16 @@ Enrolled recipes = the cc-ci `tests/<recipe>/` dirs (same set `ci-test-review` s
 ```
 ssh cc-ci 'cd /root/cc-ci/tests && ls -d */' | sed 's#/##' | grep -vE '^(_generic|unit|__pycache__)$'
 ```
-(or the names passed in `$ARGUMENTS`). For each, on cc-ci, check availability — skip dirty/up-to-date:
+(or the names passed in `$ARGUMENTS`). For each, on cc-ci, check availability — skip dirty/up-to-date.
+
+> ⚠️ **Run abra over a pseudo-TTY** or it FATAs `inappropriate ioctl for device` (no TTY under plain
+> ssh). Wrap every abra call in `script` (the working-harness fix in `runner/harness/abra.py`) and pass
+> `-n`: `ssh cc-ci 'script -qec "abra <args> -n" /dev/null'`. `git`/other commands need no wrapper.
 ```
 ssh cc-ci 'export PATH=/run/current-system/sw/bin:$PATH; \
   git -C ~/.abra/recipes/<r> status --short; \
-  abra recipe fetch <r> --force; \
-  abra recipe upgrade <r> -m -n'
+  script -qec "abra recipe fetch <r> --force -n" /dev/null; \
+  script -qec "abra recipe upgrade <r> -m -n" /dev/null'
 ```
 Build `RECIPES_TO_UPGRADE` = recipes with a **clean worktree** AND **≥1 available upgrade**. Others go
 to `SKIPPED_UPFRONT` with a reason (`dirty-worktree`, `up-to-date`, `not-fetchable`).
@@ -103,7 +107,8 @@ End with the report path and a reminder that **nothing was merged**.
 ## Safety / coordination (this matters — shared host with the build loops)
 - **Sequential is the default for a reason.** Recipe deploys are **stateful on the shared Swarm** and
   parallel deploys can OOM/collide. Between sequential recipes, the per-recipe `recipe-upgrade` tears
-  down what it deployed; verify a recipe is undeployed before the next starts (`abra app ls` on cc-ci).
+  down what it deployed; verify a recipe is undeployed before the next starts
+  (`ssh cc-ci 'script -qec "abra app ls -n" /dev/null'` — pseudo-TTY wrapped, per the box above).
 - **Single-writer:** every PR (recipe or cc-ci test) is on a dedicated branch; **never push `main`**,
   never touch the build loops' `/cc-ci` `/cc-ci-adv` working clones or their in-flight state.
 - **Contention with active loop development:** while the loops are still building cc-ci, this run