Introduction

ctk (Confluence Toolkit) is a single Go binary that serves as both a command-line interface and an MCP (Model Context Protocol) server for Confluence Cloud.

Why ctk?

ctk follows the Unix philosophy: one tool, one job. Where jtk handles Jira, ctk handles Confluence. Each is a standalone binary with its own credentials, its own MCP tools, and its own release cycle. No monoliths.

Key Features

• Dual mode — CLI for humans, MCP server for AI agents
• Storage format conversion — XHTML to markdown on the fly for LLM consumption
• Folder support — list, create, navigate, and delete Confluence Cloud folders
• Page diffing — unified diff between any two page versions
• Inline comments — read inline and footer comments with highlighted text
• Attachments — list and download page attachments
• CQL search — full Confluence Query Language plus quick text search
• Write gating — destructive operations are off by default (CTK_ENABLE_WRITES)
• Rate limiting — built-in token-bucket limiter (20 req/min) to stay under Confluence API limits

Comparison

Feature	ctk	sooperset	aashari	aaronsb
Folders	Yes	No	No	No
Page diff	Yes	No	No	No
Inline comments	Yes	No	No	No
Attachments	Yes	No	Yes	No
Labels	Yes	No	No	No
Storage format to markdown	Yes	No	No	No
Write gating	Yes	No	No	No
CLI mode	Yes	No	No	No
Response flattening	Yes	Partial	No	No
Rate limiter	Yes	No	No	No

Tip

ctk uses the Confluence Cloud V2 REST API with a V1 fallback for CQL search (the V2 API does not expose a CQL endpoint).