No Polling Agents

Stops Claude from burning tokens by obsessively polling background agents in sleep loops. When you launch parallel agents with run_in_background, Claude's instinct is often to check status files every few seconds until they complete. This skill teaches it to launch the agents and then just continue working or chatting until you actually need the results. Check status only when the user asks or when you hit a blocking dependency. Comes from user feedback that agents can just ping when they're done, so there's no reason to sit there polling like it's 2005.

skills.sh installs: 483
GitHub stars: 3.9K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add parcadei/continuous-claude-v3 --skill no-polling-agents --agent claude-code

Installs into .claude/skills of the current project.

Files

No Polling for Background Agents

When launching parallel background agents, do NOT poll with sleep loops.

Pattern

Background agents write to status files when complete. Wait for them naturally.

DO

Launch agents with run_in_background: true
Continue with other work while agents run
Check status file only when user asks or when you need results to proceed
Trust the agent completion system

DON'T

Run sleep 10 && cat status.txt in loops
Continuously poll for completion
Waste tokens checking status repeatedly
Block on agents unless absolutely necessary

When to Check Status

User explicitly asks "are they done?"
You need agent output to proceed with next task
Significant time has passed and user is waiting

Example

// Launch agents
Task({ ..., run_in_background: true })
Task({ ..., run_in_background: true })

// Continue with other work or conversation
// Agents will write to status file when done

// Only check when needed
cat .claude/cache/status.txt