General-purpose coding policy for Baruch's AI agents
74
93%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
Low
Low-risk findings worth noting
#!/usr/bin/env bash
# Shared lane classifier for fleet dedup, release snapshots and dismissals.
# Sourcing defines review_lanes_jq; execution emits the marker as JSON.
set -euo pipefail
CLAUDE_REVIEW_LOGIN="jbaruch-claude-reviewer[bot]"
review_lanes_jq() {
cat <<'JQ'
# Shared by fleet deduplication, release snapshots and stale-review dismissal.
# Legacy unmarked reviews from the trusted reviewer identities are policy reviews.
def quality_marker: "<!-- coding-policy:claude-quality-review:v1 -->";
def is_quality_review: ((.body // "") | startswith(quality_marker));
def in_review_lane($lane):
if $lane == "quality" then is_quality_review
elif $lane == "policy" then (is_quality_review | not)
else error("unknown review lane: " + $lane) end;
JQ
}
main() {
if ! command -v jq >/dev/null 2>&1; then
echo "error: jq is not installed — install it with 'brew install jq' (macOS) or 'apt install jq' (Debian/Ubuntu), then retry" >&2
return 2
fi
jq -n --arg login "$CLAUDE_REVIEW_LOGIN" "$(review_lanes_jq)"' {quality_marker: quality_marker, quality_login: $login}'
}
if [[ "${BASH_SOURCE[0]}" == "${0}" ]]; then
main "$@"
fi.tessl-plugin
rules
skills
migrate-to-plugin
onboard-repo
release
hooks
references