← Back to Skills Marketplace
dodge1218

Task Extractor

by KairoKid · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ Security Clean
151
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install dsb-task-extractor
Description
Extract, track, and verify completion of multiple tasks from a single user message. Use when a message contains 3+ actionable items or mixes multiple unrelat...
README (SKILL.md)

Task Extractor

When to Trigger

A user message contains 3+ actionable items, mixes asks across different domains, or has a history of dropped tasks.

Process

  1. Extract every task from the message — err on the side of capturing more.
  2. Number each task explicitly.
  3. Track in a hot index: | # | Task | Status | Evidence |
  4. Execute each in order (or delegate to appropriate skill).
  5. Reconcile after completion — compare extracted list against actual completions.

Example

User: "Fix the deploy bug, update the landing page copy, 
       check if Stripe webhook is working, and email the client"

Extracted:
1. Fix deploy bug → coder
2. Update landing page copy → coder
3. Check Stripe webhook → verifier
4. Email client → GATE (needs approval)

Rules

  • Missing a task is worse than capturing an extra one.
  • Mark each explicitly: ✅ done | 🔄 in progress | ❌ blocked | ⏭️ deferred.
  • At end of run, list any DROPPED tasks — these are the most important to surface.
  • If a task needs human approval, mark as GATE and don't execute silently.
Usage Guidance
This skill appears to do what it claims, but before installing verify two things: (1) how and where the "hot index" (task tracking table) will be stored or logged — the SKILL.md doesn't specify storage or retention, which could expose task contents; (2) whether the agent is allowed to automatically execute or delegate tasks in your environment — although the skill calls out GATE for approvals, execution instructions could lead to side effects (sending emails, running deployments) if not properly gated. If you plan to use it, test with non-sensitive tasks, enable strict gating for any action with potential impact (payments, messages, deployments), and confirm audit/logging so you can review what was executed.
Capability Analysis
Type: OpenClaw Skill Name: dsb-task-extractor Version: 1.0.0 The task-extractor skill bundle is a meta-instruction set designed to improve agent reliability when handling complex, multi-part user requests. It provides a structured process for task extraction, tracking, and reconciliation in SKILL.md, including safety features like human-in-the-loop requirements (GATE) for sensitive actions. No malicious code, data exfiltration, or suspicious execution patterns were identified.
Capability Assessment
Purpose & Capability
Name/description match the instructions: SKILL.md focuses on extracting, numbering, tracking, and reconciling tasks. No unrelated environment variables, binaries, or installs are requested.
Instruction Scope
Instructions are mostly scoped to extraction and tracking, but they also tell the agent to "Execute each in order (or delegate to appropriate skill)". That expands runtime behavior beyond pure extraction and could cause the agent to invoke other skills or perform actions with side effects. The SKILL.md does include a GATE rule for human approval, which limits silent execution, but storage location for the 'hot index' and limits on what can be executed are unspecified.
Install Mechanism
Instruction-only skill with no install spec or code files — minimal install risk and nothing is written to disk by the skill itself.
Credentials
No environment variables, credentials, or config paths are requested; nothing disproportionate is being asked for.
Persistence & Privilege
always:false and no declared persistence. The skill can be invoked autonomously (platform default), and its instructions to execute/delegate tasks increase the effective blast radius because it may call other skills; this is expected behavior but worth confirming with site policies or user preferences.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install dsb-task-extractor
  3. After installation, invoke the skill by name or use /dsb-task-extractor
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release: Extract and track multiple tasks from compound user messages. Prevents dropped tasks.
Metadata
Slug dsb-task-extractor
Version 1.0.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is Task Extractor?

Extract, track, and verify completion of multiple tasks from a single user message. Use when a message contains 3+ actionable items or mixes multiple unrelat... It is an AI Agent Skill for Claude Code / OpenClaw, with 151 downloads so far.

How do I install Task Extractor?

Run "/install dsb-task-extractor" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Task Extractor free?

Yes, Task Extractor is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Task Extractor support?

Task Extractor is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Task Extractor?

It is built and maintained by KairoKid (@dodge1218); the current version is v1.0.0.

💬 Comments