Does this skill work with agent swarm architectures?

Yes, it is specifically designed for orchestration or planning agents that need to produce high-quality instructions for a fleet of worker agents to execute without human intervention.

When should I use the CLEAR + CoVe skill?

Use it when you are producing task prompts for worker agents, planning multi-step autonomous workflows, or creating tasks that require high factual accuracy, such as API configurations or version-specific implementations.

What is the CLEAR framework in task design?

CLEAR stands for Concise, Logical, Explicit, Adaptive, and Reflective. It is a writing framework that ensures agent tasks are brief, logically structured, boundary-defined, flexible where necessary, and include self-validation steps.

How does Chain of Verification (CoVe) prevent hallucinations?

CoVe forces the agent to generate an initial draft, create independent verification questions to check factual claims, answer those questions using primary sources, and then revise the final output based on the findings.

CLEAR + CoVe Task Designer

Name: CLEAR + CoVe Task Designer
Author: Jamie-BitFlight

byJamie-BitFlight

•

Productivity & Workflow

Optimizes agent task instructions using the CLEAR framework and Chain of Verification to eliminate ambiguity and hallucination.

This skill provides a standardized framework for designing high-precision task files for agent swarms and worker agents. By combining the CLEAR writing methodology (Concise, Logical, Explicit, Adaptive, Reflective) with the Chain of Verification (CoVe) reliability pattern, it ensures that autonomous agents receive unambiguous, independently executable, and verifiable instructions. It is particularly useful for orchestration agents planning multi-step workflows where factual accuracy, strict acceptance criteria, and robust verification are critical to prevent project divergence and errors.

Key Features

01Implements the CLEAR framework for structured, logical task prompt design

02Supports swarm orchestration by ensuring task atomicity and clear handoff protocols

03Lints task quality against concise, explicit, and reflective criteria

04Integrates Chain of Verification (CoVe) to minimize model hallucinations in high-risk tasks

0526 GitHub stars

06Provides standardized templates for verifiable and actionable worker agent instructions

Use Cases

01Refining ambiguous project requirements into actionable, testable technical tasks for LLMs

02Generating high-fidelity TASK.md files for autonomous worker agents in a swarm

03Orchestrating complex multi-agent workflows that require strict quality gates and verification

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add jamie-bitflight/claude_skills clear-cove-task-design

For use in Claude.ai and ChatGPT

Download Skill