AgentStack
SKILL verified MIT Self-run

Jobs

skill-damon-stewart-1-claude-skills-public-jobs · by Damon-Stewart-1

Check status of dispatched background Claude tasks. Use when the user says "jobs", "check jobs", "job status", "what's running", or "/jobs".

No reviews yet
0 installs
8 views
0.0% view→install

Install

$ agentstack add skill-damon-stewart-1-claude-skills-public-jobs

✓ scanned · ✓ verified — works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

Are you the author of Jobs? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Jobs: Check Background Task Status

The user wants to see the status of dispatched background Claude tasks.

What to do

  1. List all job metadata files:
ls -t ~/.claude/jobs/*.meta 2>/dev/null

If no files exist, tell the user "No dispatched jobs found. Use /dispatch to send a task to background."

  1. For each job, read the .meta file and display a summary table:

| Job ID | Status | Task | Started | Exit Code | |--------|--------|------|---------|-----------|

Status indicators and display treatment:

  • complete (green)
  • running (blue)
  • failed (red): non-zero exit, not a timeout
  • timed_out (orange): exit code 124, gtimeout killed it; was actively running, not crashed. Offer to re-dispatch with a higher TIMEOUT_SECS.
  • stale (yellow): process died silently without updating meta; see step 5

The exit_code field may be absent in older meta files; show blank if missing.

  1. For completed jobs, check if the output file exists and offer to read it:
  • "Job X is complete. Want me to read the output?"
  1. For failed or timed_out jobs, automatically show the last 20 lines of the .log file:
tail -20 ~/.claude/jobs/.log

Include the exit_code from the meta file in the summary. If a failed job has an output file (partial results from fallback extraction), mention it.

  1. For running jobs, check if the process is still alive:
# Check if any claude process is running for this job's log file
ps aux | grep -l "claude" | head -5

If the process died but status still says "running", update the meta file to "failed" and check the log file for errors.

  1. If the user asks about a specific job, read both the .meta and .log files. The log contains the full JSON output from Claude, including session ID. If the user wants to continue a job's conversation:
unset CLAUDECODE && claude -p "Continue where you left off" --resume 

Cleanup

If the user says "clean jobs" or "clear jobs", offer to remove completed and failed job files:

# Remove .meta, .log, and .md files for completed and failed jobs
grep -l "status: \(complete\|failed\)" ~/.claude/jobs/*.meta | while read f; do
  base="${f%.meta}"
  rm -f "$f" "${base}.log" "${base}.md"
done

Always confirm before deleting.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet — be the first.

Versions

  • v0.1.0 Imported from the upstream source.