No Task Output

This prevents a sneaky context overflow issue where TaskOutput dumps entire agent transcripts (70k+ tokens) into your main conversation, triggering mid-conversation compaction and defeating the whole point of isolated agent contexts. It enforces using Task in synchronous mode instead, which keeps the agent work contained and just returns a summary. Born from an actual session where someone got bitten by this. If you're running background tasks and wondering why your context suddenly exploded, this is your guardrail. Simple rule, solves a real footgun.

skills.sh installs: 481
GitHub stars: 3.9K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add parcadei/continuous-claude-v3 --skill no-task-output --agent claude-code

Installs into .claude/skills of the current project.

Files

Never Use TaskOutput

TaskOutput floods the main context window with agent transcripts (70k+ tokens).

Rule

NEVER use TaskOutput tool. Use Task tool with synchronous mode instead.

Why

TaskOutput reads full agent transcript into context
This causes mid-conversation compaction
Defeats the purpose of agent context isolation

Pattern

# WRONG - floods context
Task(run_in_background=true)
TaskOutput(task_id="...")  // 70k tokens dumped

# RIGHT - isolated context, returns summary
Task(run_in_background=false)  // Agent runs, returns summary