Chain of Verification

Apply Chain-of-Verification (CoVe) prompting to improve response accuracy through self-verification. Use when complex questions require fact-checking, technical accuracy, or multi-step reasoning.

155 stars 30 forksAI, LLMs & Agents Plugin marketplace

Install

Claude Code plugin
/plugin marketplace add serpro69/claude-toolbox

The repo is a Claude Code plugin marketplace. Run /plugin afterwards to pick the plugin that includes this skill.

skills CLI (Claude Code, Cursor, Codex and more)
npx skills add serpro69/claude-toolbox --skill chain-of-verification
Manual install
git clone --depth 1 https://github.com/serpro69/claude-toolbox.git /tmp/claude-toolbox
mkdir -p ~/.claude/skills
cp -r /tmp/claude-toolbox/klaude-plugin/skills/chain-of-verification ~/.claude/skills/chain-of-verification

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# Chain-of-Verification (CoVe)

CoVe is a verification technique that improves response accuracy by making the model fact-check its own answers. Instead of accepting an initial response at face value, CoVe instructs the model to generate verification questions, answer them independently, and revise the original answer based on findings.

## Conventions

- **Read capy knowledge base conventions** at [shared-capy-knowledge-protocol.md](shared-capy-knowledge-protocol.md).

**Capy restriction:** CoVe is a read-only verification tool. Do NOT call `capy_index` or `capy_fetch_and_index` during this workflow. Use `capy_search` only. If corrections reveal knowledge worth persisting, the calling agent handles indexing after CoVe completes.

## When to Use This Skill

CoVe adds the most value in these scenarios:

**Precision-required questions:**

- Questions containing precision language ("exactly", "precisely", "specific")
- Complex factual questions (dates, statistics, specifications)

**Complex reasoning:**

- Multi-step reasoning chains (3+ logical dependencies)
- Technical claims about APIs, libraries, or version-specific behavior

**Fact-checking scenarios:**

- Historical facts, statistics, or quantitative data
- Technical specifications and API behavior

**High-stakes accuracy:**

- Security-critical code paths or analysis
- Code generation requiring accuracy verification
- Any response where correctness is critical

**Self-correction triggers:**
Read the full skill on GitHub

Details

Skill name
chain-of-verification
Path
klaude-plugin/skills/chain-of-verification/SKILL.md
License
Apache-2.0
Repo updated
18 November 2026

FAQ

What does the Chain of Verification skill do?

Apply Chain-of-Verification (CoVe) prompting to improve response accuracy through self-verification. Use when complex questions require fact-checking, technical accuracy, or multi-step reasoning.

How do I install the Chain of Verification skill?

Claude Code plugin: /plugin marketplace add serpro69/claude-toolbox. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Chain of Verification popular?

It lives in serpro69/claude-toolbox, which has 155 GitHub stars and 30 forks, and is included in 1 curated skill list.

Related skills

Brainstorm

155

serpro69/claude-toolbox

TRIGGER when: asked to brainstorm, think through, explore, or pressure-test a technical idea or decision through conversation. Interviews adaptively, asks one question at a time, and closes with a chat recap. For technical products, architecture, infrastructure, tools, operations, and engineering workflows. Written design/specification and task-list requests belong to /kk:design; durable domain glossary/reference kits to /kk:model; implementation and fixes to /kk:implement. Not for nontechnical brainstorming.

Productivity & NotesIn 1 list30

Clarify Docs

155

serpro69/claude-toolbox

TRIGGER when: asked to clarify or improve existing documentation, PR drafts or issue descriptions, including URLs and pasted bodies. Saves results locally, grounded in requirements and source while preserving meaning. Not for requests to implement, fix or work on an issue, code, config, generic response brevity, or agent instructions. Explicit instruction or SKILL.md targets receive a /kk:implement suggestion without edits or invoking implementation.

Writing & ContentIn 1 list30

Dependency Handling

155

serpro69/claude-toolbox

TRIGGER when: adding or upgrading any dependency — library, SDK, framework, API, IaC API version (K8s/Terraform/Helm), CRD, or container image. Use BEFORE writing the call. Forces context7/capy lookup instead of guessing.

DevOps & CloudIn 1 list30

LLM Wiki Setup

103

NulightJens/ai-second-brain-skills

Initialize a Karpathy-style LLM wiki (personal knowledge base / AI second brain) in a folder. Creates a CLAUDE.md schema with routing table and naming conventions, AGENTS.md mirror, raw/wiki directory structure, and index.md/log.md scaffolding — a complete working vault in one pass. Use when the user asks to "set up an LLM wiki", "create a second brain", "build a personal knowledge base", "install the Karpathy wiki pattern", "initialize a knowledge vault", or points at an empty folder and wants to turn it into an AI-maintained wiki.

AI, LLMs & Agents17

Awesome Gaussian Skills

188

jaccen/Awesome-Gaussian-Skills

Umbrella router for the Awesome-Gaussian-Skills toolkit (github.com/jaccen/Awesome-Gaussian-Skills): 1011 indexed 3D Gaussian Splatting methods (904 with an arXiv ID) across 26 categories, 18 specialist sub-skills, a role-based research expert team (科研专家团, evidence-led review), and a spatial-intelligence / world-model / embodied-AI knowledge base. Routes idea evaluation, contribution analysis, critical review, editing, publication integrity, venue checks, experiment design and result analysis to the right role on demand; also handles method survey, citation verification and benchmark lookup. U

AI, LLMs & Agents13

Model

155

serpro69/claude-toolbox

Produce or update a domain-reference kit — the durable glossary + divergences/traps pages for a bounded context. Use for domain modelling, information architecture, building a domain glossary or domain model, mapping domain concepts to code (bindings), or surfacing the decidable questions stakeholders must close before design. Runs upstream of feature /kk:design as the first producer of the architecture flow, and emits artifacts /kk:review-architecture verifies. Triggers: "model the domain", "build a glossary", "domain reference", "bounded context", "what are the concepts here". NOT for behavi

AI, LLMs & AgentsIn 1 list30

Browse the most popular Claude skills, all ai, llms & agents skills, or write your own with the skills generator.