Applies Chain-of-Verification (CoVe) prompting to self-verify responses and improve accuracy for complex questions, fact-checking, technical specs, and multi-step reasoning.
From kknpx claudepluginhub serpro69/claude-toolbox --plugin kkThis skill uses the workspace's default tool permissions.
cove-isolated.mdcove-process.mdSearches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Searches prompts.chat for AI prompt templates by keyword or category, retrieves by ID with variable handling, and improves prompts via AI. Use for discovering or enhancing prompts.
Executes pre-written implementation plans: critically reviews, follows bite-sized steps exactly, runs verifications, tracks progress with checkpoints, uses git worktrees, stops on blockers.
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.
CoVe adds the most value in these scenarios:
Precision-required questions:
Complex reasoning:
Fact-checking scenarios:
High-stakes accuracy:
Self-correction triggers:
Note: These heuristics can be copied to your project's CLAUDE.md if you want Claude to auto-invoke CoVe for matching scenarios. By default, CoVe requires manual invocation to give you control over when to invest additional tokens/time for verification.
CoVe offers two verification modes to balance accuracy vs. cost:
/cove)Uses prompt-based isolation within a single conversation turn.
See cove-process.md for the standard workflow.
/kk:cove:cove-isolated)Uses Claude Code's Task tool to spawn isolated sub-agents for true factored verification.
Sub-agent customization flags:
| Flag | Effect |
|---|---|
--explore | Use Explore agent for codebase verification |
--haiku | Use haiku model for faster/cheaper verification |
--agent=<name> | Use custom agent type |
See cove-isolated.md for the isolated workflow.
| Use Case | Recommended Mode |
|---|---|
| Quick fact-checking | /cove |
| High-stakes accuracy | /kk:cove:cove-isolated |
| Codebase verification | /kk:cove:cove-isolated --explore |
| Cost-sensitive verification | /cove or /kk:cove:cove-isolated --haiku |
The CoVe workflow follows 4 steps:
See cove-process.md for the standard workflow, or cove-isolated.md for the isolated sub-agent workflow.
Use the /cove skill followed by your question:
/cove What is the time complexity of Python's sorted() function?
Or invoke /cove after receiving a response to verify it.
For isolated verification with sub-agents:
/kk:cove:cove-isolated What is the time complexity of Python's sorted() function?
With flags:
/kk:cove:cove-isolated --explore How does the auth system work?
/kk:cove:cove-isolated --haiku What year was TCP standardized?
Claude should recognize these phrases as requests to invoke the CoVe skill:
For isolated mode:
Important: This is guidance for manual recognition only. Auto-trigger is NOT implemented by default per design goals. Users who want automatic CoVe invocation for certain scenarios can add the heuristics from "When to Use This Skill" to their project's CLAUDE.md.