From 00fb74c76a9bc1ab000e68500ed06c8f6efdaf64 Mon Sep 17 00:00:00 2001
From: Matteo Cherubini <cherubinimatte@gmail.com>
Date: Fri, 5 Jun 2026 09:59:18 +0200
Subject: [PATCH] feat(ingest): Improve general script robustness and cleanup

---
 skills/ingest/scripts/log-append.sh  |  2 +-
 skills/ingest/scripts/run-ingest.sh  | 13 ++++++++++++-
 skills/ingest/scripts/scoped-lint.sh |  7 ++++++-
 3 files changed, 19 insertions(+), 3 deletions(-)

diff --git a/skills/ingest/scripts/log-append.sh b/skills/ingest/scripts/log-append.sh
index 32e6ca0..b3108a2 100644
--- a/skills/ingest/scripts/log-append.sh
+++ b/skills/ingest/scripts/log-append.sh
@@ -35,7 +35,7 @@ esac
 
 [[ -f "$LOG_FILE" ]] || { echo "log-append: not found: $LOG_FILE" >&2; exit 1; }
 
-run_id="$(uuidgen 2>/dev/null || cat /proc/sys/kernel/random/uuid)"
+run_id="$(uuidgen 2>/dev/null || cat /proc/sys/kernel/random/uuid 2>/dev/null || python3 -c 'import uuid; print(uuid.uuid4())')"
 today="$(date +%Y-%m-%d)"
 
 {
diff --git a/skills/ingest/scripts/run-ingest.sh b/skills/ingest/scripts/run-ingest.sh
index 032761c..4ffbd1f 100644
--- a/skills/ingest/scripts/run-ingest.sh
+++ b/skills/ingest/scripts/run-ingest.sh
@@ -46,6 +46,11 @@ all_paths=( "${created_paths[@]}" "${modified_paths[@]}" )
 
 conflict_label=""
 
+# NOTE: no rollback. Steps below mutate the working tree in order (index → log → commit).
+# All are idempotent on re-run EXCEPT log-append (append-only). If a step fails midway,
+# nothing is committed (open-pr is the only committer) — the operator re-runs, or inspects
+# wiki/ if log-append already wrote a line. The manifest is removed only on full success.
+
 # --- 1. index entries (created pages only), inserted in order ---
 while IFS=$'\t' read -r path summary maturity; do
   [[ -z "$path" ]] && continue
@@ -119,4 +124,10 @@ jq -nc \
   --arg detail   "$pr_out" \
   '{status:$status, slug:$slug, pr_url:$pr_url, lint_clean:$lint_clean, conflict:$conflict, detail:$detail}'
 
-[[ $pr_rc -eq 0 ]]
+# The manifest is a single file overwritten by each pi run (not accumulating), but on full
+# success we remove it so a stale manifest can never be re-processed by mistake.
+if [[ $pr_rc -eq 0 ]]; then
+  rm -f "$manifest"
+else
+  exit 1
+fi
diff --git a/skills/ingest/scripts/scoped-lint.sh b/skills/ingest/scripts/scoped-lint.sh
index a064fd9..ded50a1 100644
--- a/skills/ingest/scripts/scoped-lint.sh
+++ b/skills/ingest/scripts/scoped-lint.sh
@@ -12,7 +12,12 @@
 # =============================================================================
 set -euo pipefail
 
-: "${KG_LIB_DIR:?set KG_LIB_DIR to the framework lib/ dir (e.g. /opt/knowledge-genome-setup/lib)}"
+: "${KG_LIB_DIR:?set KG_LIB_DIR to the framework lib/ dir (e.g. /opt/knowledge-genome-orchestrator/lib)}"
+
+# Fail clearly if the lib files are missing, rather than a raw `source: No such file`.
+for _f in output.sh lint.sh; do
+  [[ -f "${KG_LIB_DIR}/${_f}" ]] || { echo "scoped-lint: missing ${KG_LIB_DIR}/${_f}" >&2; exit 1; }
+done
 
 # shellcheck source=/dev/null
 source "${KG_LIB_DIR}/output.sh"