Menu
Run multiple commands in parallel and aggregate outputs. Load when tasks are independent and can run concurrently.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Use when opening a PR and driving it to green CI — push the branch, write a high-level PR description, watch checks, fix failures, and keep looping until green; also use when an already-open PR gets new commits.
Use when the user starts, resumes, switches, saves, or recalls a named task or ongoing work across sessions (e.g. "remember this", "continue X", "what was I doing on Y", "track this task").
Git commit workflow. Load when finishing any code-change task (after verification) to commit and push, or when explicitly staging/committing.
Use when modifying or adding a transform, loss, model, method, package, task-model, config, docs page, or example in a Lightly AG repo (lightly-ssl or lightly-train) and you need repo-specific conventions.
Spawn a subagent to run a task. Use when you want to delegate work to a separate pi instance.
Use when configuring or replicating LT-DETR v2 (ltdetrv2-s/m/l/x) COCO benchmarks via the lightly-train wrapper — model alias map, global vs per-rank batch size, recipe invariants, ECDet config mapping, cluster pins, and common gotchas.
| name | parallel |
| description | Run multiple commands in parallel and aggregate outputs. Load when tasks are independent and can run concurrently. |
Run independent commands concurrently for efficiency.
mkdir -p /tmp/parallel-work
(ddgs text -q "query1" -m 15 > /tmp/parallel-work/01.txt) &
PID1=$!
(ddgs text -q "query2" -m 15 > /tmp/parallel-work/02.txt) &
PID2=$!
(ddgs text -q "query3" -m 15 > /tmp/parallel-work/03.txt) &
PID3=$!
(ddgs text -q "query4" -m 15 > /tmp/parallel-work/04.txt) &
PID4=$!
wait $PID1 $PID2 $PID3 $PID4
cat /tmp/parallel-work/0*.txt
mkdir -p /tmp/my-temp
(cmd1 > /tmp/my-temp/r1.txt) &
PID1=$!
(cmd2 > /tmp/my-temp/r2.txt) &
PID2=$!
(cmd3 > /tmp/my-temp/r3.txt) &
PID3=$!
wait $PID1 $PID2 $PID3
cat /tmp/my-temp/r*.txt
mkdir -p /tmp/parallel-work
PIDS=()
for item in "${ITEMS[@]}"; do
(process "$item" > "/tmp/parallel-work/r_$item.txt") &
PIDS+=($!)
done
wait "${PIDS[@]}"
cat /tmp/parallel-work/r_*.txt
(timeout 60s cmd > /tmp/my-temp/result.txt) &
(cmd > /tmp/my-temp/r.txt) &
PID=$!
if wait $PID; then
echo "Success"
cat /tmp/my-temp/r.txt
else
echo "Failed with exit code: $?"
fi
| Guideline | Reason |
|---|---|
| Use 4-6 concurrent | Sweet spot for most systems |
| Unique temp dir per run | Avoids file collisions |
| Use absolute paths | Avoids cwd confusion |
Always wait | Ensures all processes finish |
| Parenthesize each bg job | (cmd &) not { cmd & } |
⚠️ Critical: Use subshells for background jobs
Pattern (command > output.txt) & works reliably.
Pattern { command > output.txt & } does NOT work with ddgs.
⚠️ Danger: Temp file race conditions
Never use $$ or variable temp paths in filenames across parallel invocations.