General-purpose coding policy for Baruch's AI agents
73
92%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
View guide
Low
Low-risk findings worth noting
#!/usr/bin/env bash
# Start and verify the judge tier recorded by teamlead plan.
# Contract: <plan-file> <pane> [claude|codex|grok] [--state FILE] [--task TASK] [--now ISO].
# Requires an empty shell pane; use apply for an existing worker's relaunch.
# stdout: JSON {agent, model, effort, pane, argv_verified, verified} on success.
# Exit 0: launch arguments proved the tier; 1: input/transport/proof failure;
# 2: command-line usage error. No banner or transcript text establishes proof.
# HERDR_BIN overrides the transport; Python selection is owned by teamlead.sh.
set -euo pipefail
main() {
if (( $# < 2 )); then
echo "start-judge-worker: usage: start-judge-worker.sh <plan-file> <pane> [kind] [--state FILE] [--task TASK] [--now ISO]" >&2
return 2
fi
local skill_dir
skill_dir="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
local plan="$1" pane="$2" kind="claude"
shift 2
if (( $# > 0 )) && [[ "$1" != --* ]]; then
kind="$1"
shift
fi
bash "${skill_dir}/teamlead.sh" start-judge --assignments "$plan" --pane "$pane" \
--kind "$kind" --herdr-bin "${HERDR_BIN:-herdr}" "$@"
}
if [[ "${BASH_SOURCE[0]}" == "$0" ]]; then
main "$@"
fi.tessl-plugin
hooks
rules
skills
adopt-fork-pr
herdr-standup
herdr-teamlead
references
teamlead
templates
tests
migrate-to-plugin
onboard-repo
release
tests