Skip to content

Business operating system for Claude Code — 57 skills, 21 agents, smart daemon. Unified inbox (WhatsApp/Email/Slack/Telegram), autonomous PR merge, full-AWS monitoring, revenue (Stripe+RevenueCat), e-commerce (Shopify), marketing (Klaviyo/Meta/GA4), voice (Bland/ElevenLabs), APM (Datadog/NewRelic/OTEL), YOLO mode.

fix(hooks): scan only added lines in secret hook, check all rm targets

auroracapital7e1d6ddf950e
diff --git a/claude-ops/bin/ops-no-rm-rf-anchor b/claude-ops/bin/ops-no-rm-rf-anchor
index 670c0920..63f8c164 100755
--- a/claude-ops/bin/ops-no-rm-rf-anchor
+++ b/claude-ops/bin/ops-no-rm-rf-anchor
@@ -38,7 +38,7 @@ echo "$CMD" | grep -Eq '(^|[[:space:];&|`])rm([[:space:]]|$)' || exit 0
NORM="$(echo "$CMD" | tr '\n' ' ' | tr -s ' ')"
# Match `rm` with -rf / -fr / -r -f (any combo of -r, -R, -f, --recursive, --force).
-# Then check the FIRST argument after the flags for dangerous anchors.
+# Then check ALL arguments after the flags for dangerous anchors.
#
# Dangerous anchor regex matches the target token explicitly:
# / /* /.
@@ -100,8 +100,7 @@ DANGER=$(set -f; while IFS= read -r seg; do
if [[ "$tok" == '"$HOME"' || "$tok" == '"$HOME"/' || "$tok" == '"$HOME"/*' ]]; then
DANGER="$tok"; break
fi
- # Only inspect the first non-flag target — that's enough to flag.
- break
+ # Continue checking remaining targets — any dangerous anchor in the list blocks.
done
if [[ -n "$DANGER" ]]; then
diff --git a/claude-ops/bin/ops-prevent-secret-commit b/claude-ops/bin/ops-prevent-secret-commit
index 6442f6a8..2e6c7c85 100755
--- a/claude-ops/bin/ops-prevent-secret-commit
+++ b/claude-ops/bin/ops-prevent-secret-commit
@@ -24,7 +24,12 @@ command -v git >/dev/null 2>&1 || exit 0
git rev-parse --is-inside-work-tree >/dev/null 2>&1 || exit 0
# Capture staged diff (cwd-relative). Limit size to keep <100ms.
-DIFF="$(git diff --cached --no-color 2>/dev/null | head -c 524288 || true)"
+FULL_DIFF="$(git diff --cached --no-color 2>/dev/null | head -c 524288 || true)"
+[[ -z "$FULL_DIFF" ]] && exit 0
+
+# Only scan added lines (+) to avoid false positives when removing secrets.
+# Keep +++ headers for file attribution, strip --- headers and context lines.
+DIFF="$(echo "$FULL_DIFF" | grep -E '^(\+|\+\+\+ b/)' | grep -v '^--- ' || true)"
[[ -z "$DIFF" ]] && exit 0
# Patterns: label|regex (POSIX ERE, BSD-grep compatible — no \d, no \b lookarounds).