Background Agent Pings

by parcadeid07ff4b06b62No license3.9K starsListed Oct 8, 2026Updated Oct 8, 2026Repository updated 8 months ago

Instructions onlyAI & Agents

AI-generated overview

Guides agents to rely on system reminders for background agent progress instead of polling for output.

What it does: This skill describes a workflow pattern for handling background agents: launch the agent with an output file path, continue other work immediately, and wait for system reminders about agent progress. It advises checking the output file only when the agent signals completion, and lists polling behaviors such as sleep-and-list, tailing task output, or calling TaskOutput as things to avoid. It explains that polling wastes tokens and that TaskOutput floods the main context with the full agent transcript.
When to use it: Use it when launching background agents and deciding how to monitor their progress. It fits situations where an agent runs asynchronously and you want to avoid token-wasting polling. It is not user-invocable and is meant as internal guidance for the agent.
Requirements: No scripts or tools beyond the agent environment; it assumes background agent launching and system reminders are available.

Background Agent Pings

Trust system reminders as agent progress notifications. Don't poll.

Pattern

When you launch a background agent, continue working on other tasks. The system will notify you via reminders when:

Agent makes progress: Agent <id> progress: X new tools used, Y new tokens
Agent writes output file (check the path you specified)

DO

1. Task(run_in_background=true, prompt="... Output to: .claude/cache/agents/<type>/output.md")2. Continue with next task immediately3. When system reminder shows agent activity, check if output file exists4. Read output file only when agent signals completion

1. Task(run_in_background=true, prompt="... Output to: .claude/cache/agents/<type>/output.md")2. Continue with next task immediately3. When system reminder shows agent activity, check if output file exists4. Read output file only when agent signals completion

DON'T

# BAD: Polling wastes tokens and timeTask(run_in_background=true)Bash("sleep 5 && ls ...")  # pollingBash("tail /tmp/claude/.../tasks/<id>.output")  # pollingTaskOutput(task_id="...")  # floods context

# BAD: Polling wastes tokens and timeTask(run_in_background=true)Bash("sleep 5 && ls ...")  # pollingBash("tail /tmp/claude/.../tasks/<id>.output")  # pollingTaskOutput(task_id="...")  # floods context

Why This Matters

Polling burns tokens on repeated checks
TaskOutput floods main context with full agent transcript
System reminders are free - they're pushed to you automatically
Continue productive work while waiting

Source

This session: Realized polling for agent output wasted time when system reminders already provide progress updates

Source and attribution

Source:parcadei/continuous-claude-v3in.claude/skills/background-agent-pingsat commitd07ff4b

License: No license

Content belongs to its original authors. SourceWeft indexes it from a public repository.

Report or request removal

Background Agent Pings

Background Agent Pings

Pattern

DO

DON'T

Why This Matters

Source

Source and attribution

More from parcadei/continuous-claude-v3

Tldr Stats

Wiring

Workflow Router

Validate Agent

Tour

Tldr Overview

More in AI & Agents

Ppt Template Creator

Skill Development

Plugin Structure

Command Development

Claude Md Improver

Find Skills