Task Understanding — Productivity skill for Claude Code
Task Understanding - Parsing Complex User Requests.
How to install Task Understanding
Installs to ~/.claude/skills/yrzhe-telegram-claude-bot-task_understanding/SKILL.md
mkdir -p ~/.claude/skills/yrzhe-telegram-claude-bot-task_understanding && curl -fsSL https://raw.githubusercontent.com/Yrzhe/telegram-claude-bot/HEAD/prompts/task_understanding.md -o ~/.claude/skills/yrzhe-telegram-claude-bot-task_understanding/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Task Understanding does
Task Understanding - Parsing Complex User Requests
Time-Bound Task Patterns
When users request tasks with time qualifiers, understand the TRUE intent:
Pattern 1: "时间 + 提醒我 + 动作"
Examples:
- "晚上提醒我调研XX"
- "明天帮我查一下XX"
- "下午提醒我分析XX"
**Correct interpretation**: Create a SCHEDULED TASK that:
- Executes the action (研究/调研/查/分析) at the specified time
- Sends results to user when done
**Wrong interpretation**:
- ❌ Execute NOW + set a separate "提醒" reminder
- ❌ Split into two independent
Alternatives in Productivity
- Nosqli — NoSQL injection scanner (MongoDB/Mongoose/operator-injection DBs) — auth bypass via $ne/$gt operators, bracket 4.5k ★
- MathModel Skill — Agent-native mathematical modeling workflow skills for Trae, Claude Code, and Codex, covering problem parsing 345 ★
- Toolset — Built-in tools are available when the user asks for calculations, conversions, local date/time utilities, URL 250 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Compact.Prompt
Your task is to create a detailed summary of the conversation so far, paying close attention to the user's exp
Agent Tmux Notify
A tmux-based monitor that watches Claude Code CLI instances across panes and automatically pops up an interact
Task Decomposer
Force a plan-and-confirm step before any multi-file or multi-step change. Invoke whenever the user requests a
Main Agent
You are a research orchestrator. You analyze user requests, delegate tasks to specialized subagents, and synth
Gaia Plan
Plan a complex feature using GAIA's task-orchestration pattern, structures the work into fresh-context subagen
Team Orchestrate
Orchestrate a complex multi-step task using Agent Teams for parallel phases
Related Agents
Planner
Use this agent when the user presents a complex, multi-step task that requires clarification, decomposition, o
Aro:Agent Sonnet
Generic task execution agent using Claude 4 Sonnet model for balanced performance. Ideal for complex reasoning
Workflow Coordinator
Central orchestrator for multi-agent workflows and complex multi-domain requests. Use when a task spans multip