Skip to main content
skill CX Codex

confluence · odyssey4me — Claude Code skill: what it does, how to install

Searches spaces via CQL, reads and edits Confluence Cloud pages using Markdown, and handles space permissions directly inside agent sessions.

MIT 198 installs ★ 13 updated 2026-10-25

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add odyssey4me/agent-skills --skill confluence -a claude-code
Open on GitHub

What confluence does

This skill connects your coding agent to Confluence Cloud through the $SKILL_DIR/scripts/confluence.py script. It handles CQL search queries, pulls pages converted into Markdown alongside referenced images, checks revision history, moves documents across the hierarchy, and sends deleted pages to trash.

Beyond basic page manipulation, the tool manages space settings and permission tables for individual users or groups. The --frontmatter export flag outputs metadata such as labels, parent ID, and space keys directly into YAML headers, enabling round-trip edits from local files.

Support is strictly limited to Confluence Cloud, so on-premise Confluence Server and Data Center installations will fail. Remote execution requires outbound HTTPS access to your Atlassian host plus an Atlassian account email and API token.

What it helps with

  • search documentation spaces using CQL filters
  • download Confluence pages to local Markdown with saved images
  • publish local markdown files directly to team spaces
  • audit and update user permissions for spaces

Example prompts for the agent

  • Search Confluence for all pages in space DEMO containing login in the title
  • Download page 123456 as Markdown with frontmatter to docs.md
  • Update Confluence page 123456 using content from updated.md

How to install confluence

1Skills CLI (Vercel)

npx skills add odyssey4me/agent-skills --skill confluence -a claude-code

2Manually: put the files in place

~/.claude/skills/confluence/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/confluence/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about confluence

How to install confluence in Claude Code?

Run npx skills add odyssey4me/agent-skills --skill confluence -a claude-code in your terminal.

What does the confluence skill do?

It lets your agent query Confluence with CQL, manage space permissions, and read or publish pages formatted as Markdown.

Does this skill support on-premise Confluence Server?

No, it supports Confluence Cloud only and authenticates via an Atlassian account email and API token.

Source text

SKILL.md · MIT
---
name: confluence
description: Search and manage Confluence pages and spaces using CQL, read/create/update pages with Markdown support. Use when working with Confluence documentation.
metadata:
  author: odyssey4me
  version: "3.0.0"
  category: documentation
  tags: "wiki, pages, spaces"
  complexity: standard
license: MIT
---

# Confluence

Interact with Confluence for content search, viewing pages, and space management.

> **Creating/Updating Content?** See [references/creating-content.md](references/creating-content.md) for page creation and updates with Markdown.

## Network access

Confluence API commands require outbound HTTPS to the configured Confluence host. Remote operations require outbound network access. If the agent environment restricts it, obtain access through that environment's approval or network controls before the first remote command, reusing authorization already granted. Local help and file processing do not need network access. Skill instructions do not grant network permission. Report DNS, connection, and timeout failures as network errors; do not reset credentials to resolve them. If approved access also fails, stop and report it. Authentication and permission errors require user action.

## Resolving script paths

`SKILL_DIR` in the examples is the directory containing this loaded `SKILL.md`.
Set it to that absolute path before running commands; do not assume the agent
sets this shell variable. Quote the script path if it contains spaces.

## Setup

This skill supports **Confluence Cloud only**. Installation and configuration
defaults are documented in [references/setup.md](references/setup.md).

Verify the setup with:

```bash
$SKILL_DIR/scripts/confluence.py check
```

## Authentication

Confluence Cloud authenticates with an Atlassian account email and API token.
See [Authentication](references/setup.md#authentication) for environment
variables, the config file and keyring storage.

## Commands

See [permissions.md](references/permissions.md) for read/write classification of each command.

### check

See [Setup Verification](references/setup.md#setup-verification) for dependency,
authentication, and connectivity checks.

### search

Search for content using CQL (Confluence Query Language).

```bash
# Basic search
$SKILL_DIR/scripts/confluence.py search "type=page AND space = DEMO"
$SKILL_DIR/scripts/confluence.py search "title~login" --space DEMO

# Filter by type
$SKILL_DIR/scripts/confluence.py search "space = DEMO" --type page

# Limit results
$SKILL_DIR/scripts/confluence.py search "type=page" --max-results 10
```

**Arguments:**
- `cql`: CQL query string (required)
- `--max-results`: Maximum number of results (default: 50)
- `--type`: Content type filter (page, blogpost, comment)
- `--space`: Limit to specific space

**See also**: [CQL Reference](#cql-reference) for query syntax

### page get

Get page content by ID or title.

```bash
# Get by title (returns Markdown by default)
$SKILL_DIR/scripts/confluence.py page get "My Page Title"

# Get by ID
$SKILL_DIR/scripts/confluence.py page get 123456

# Get without body content
$SKILL_DIR/scripts/confluence.py page get "My Page" --no-body

# Get in original format (not Markdown)
$SKILL_DIR/scripts/confluence.py page get "My Page" --raw

# Save to file with images downloaded to sibling directory
$SKILL_DIR/scripts/confluence.py page get "My Page" -o my-page.md

# Export with YAML frontmatter (for round-tripping)
$SKILL_DIR/scripts/confluence.py page get 123456 --frontmatter -o page.md
```

**Output**: Page metadata and content as Markdown. Images are downloaded to a sibling directory when using `--output`/`-o`.

**Arguments:**
- `page_identifier`: Page ID or title (required)
- `--markdown`: Output body as Markdown (default)
- `--raw`: Output in original format
- `--no-body`: Don't include body content
- `--frontmatter`: Output as markdown with YAML frontmatter (title, space, labels, parent) for round-tripping with `page create`/`page update`
- `--output`/`-o`: Write output to file; images are downloaded to a sibling directory named after the file stem

### page history

Show version history for a page.

```bash
# By title
$SKILL_DIR/scripts/confluence.py page history "My Page Title"

# By ID, limit results
$SKILL_DIR/scripts/confluence.py page history 123456 --max-results 10
```

**Arguments:**
- `page_identifier`: Page ID or title (required)
- `--max-results`: Maximum versions to return (default: 25)

### page create / update

Read [references/creating-content.md](references/creating-content.md) when creating or
updating pages for Markdown, images, frontmatter, table of contents, and internal links.

```bash
$SKILL_DIR/scripts/confluence.py page create --space DEMO --title "Documentation" \
  --body-file README.md
$SKILL_DIR/scripts/confluence.py page update 123456 --body-file updated.md
```

### page move

Move a page under a new parent, or to the space root.

```bash
# Move under a new parent
$SKILL_DIR/scripts/confluence.py page move 123456 --parent 789012

# Move to space root (no parent)
$SKILL_DIR/scripts/confluence.py page move 123456
```

### page delete

Delete a page by ID (moves to trash).

```bash
$SKILL_DIR/scripts/confluence.py page delete 123456
```

### space

Manage spaces.

```bash
# List all spaces
$SKILL_DIR/scripts/confluence.py space list

# List with limit
$SKILL_DIR/scripts/confluence.py space list --max-results 10

# Filter by type
$SKILL_DIR/scripts/confluence.py space list --type global

# Get space details
$SKILL_DIR/scripts/confluence.py space get DEMO
```

**Arguments:**
- `list`: List spaces
  - `--type`: Filter by type (global, personal)
  - `--max-results`: Maximum results
- `get <space-key>`: Get space details

For creating spaces, see [references/creating-content.md](references/creating-content.md).

### space permissions

View, add, and remove space permissions.

```bash
# List all permissions for a space
$SKILL_DIR/scripts/confluence.py space permissions list DEMO

# Filter by subject type
$SKILL_DIR/scripts/confluence.py space permissions list DEMO --subject-type group

# Add a permission
$SKILL_DIR/scripts/confluence.py space permissions add DEMO \
  --subject-type user --subject "5a1234abc" --operation read --target space

# Remove a permission by ID
$SKILL_DIR/scripts/confluence.py space permissions remove DEMO --id 2154
```

**Arguments:**
- `list <space-key>`: List permissions
  - `--subject-type`: Filter by user or group
- `add <space-key>`: Add a permission
  - `--subject-type`: user or group (required)
  - `--subject`: User account ID or group name/ID (required)
  - `--operation`: read, create, delete, export, administer, archive, restrict_content (required)
  - `--target`: space, page, blogpost, comment, attachment (required)
- `remove <space-key>`: Remove a permission
  - `--id`: Permission ID (required)

### config

Show configuration and defaults.

```bash
# Show all configuration
$SKILL_DIR/scripts/confluence.py config show

# Show space-specific defaults
$SKILL_DIR/scripts/confluence.py config show --space DEMO
```

This displays:
- Authentication settings (with masked token)
- Default CQL scope, max results, and default space
- Space-specific defaults for parent pages and labels

## CQL Reference

See the [Confluence CQL documentation](https://support.atlassian.com/confluence-cloud/articles/advanced-searching-using-cql/) for full CQL syntax.

Common patterns: `type=page`, `space=KEY`, `title~text`, `text~keyword`, `created >= now("-7d")`, `label=name`.

Combine with `AND`, `OR`, and `ORDER BY`:
```bash
$SKILL_DIR/scripts/confluence.py search "type=page AND space=DEMO ORDER BY created DESC"
```

## Examples

Use the examples under [search](#search), [page get](#page-get), and
[space](#space) for common read operations. For page creation, updates, and
batch workflows, see [Creating Content](references/creating-content.md#advanced-examples).

## Model Guidance

This skill makes API calls requiring structured input/output. A standard-capability model is recommended.

© odyssey4me · MIT · odyssey4me/agent-skills

Similar skills

All skills for Writing and docs →

confluence by other authors comes first.

confluence membrane A skill for Claude Code, Codex, Cursor, GitHub Copilot, and other platforms that connects Confluence through the Membrane CLI. Membrane handles auth and token refresh, so the agent goes straight to working with documents. CX Codex +1 2,100installs confluence-expert borghei confluence-expert is a Claude Code skill that audits and designs Confluence spaces by checking page hierarchy, labels, permissions, ownership and stale content. It runs when building or restructuring a wiki. CX Codex +1 131installs confluence-automation BuildWithClaude Community confluence-automation is a skill that creates, updates, and searches Confluence pages, and manages spaces, labels, and page hierarchies via Rube MCP. CX Codex +1 ★ 3,785repo stars confluence-assistant Waldemar Neto - github.com/waldemarnt confluence-assistant operates Confluence spaces through Atlassian MCP tools. Helps Claude Code search team documentation, create structured pages, and update existing entries using Markdown. CX Codex +1 ★ 7,191repo stars confluence-to-markdown iurykrieger confluence-to-markdown is a Claude Code skill that reads a Confluence page and returns its content as Markdown. It runs as part of /bedrock:learn and /bedrock:sync. CX Codex +1 ★ 111repo stars atlassian-confluence mcparmory atlassian-confluence connects AI coding assistants to Confluence spaces to search docs, read attachments, and draft pages. Generated from the OpenAPI specification with automated tool testing. CX Codex +1 ★ 32repo stars confluence-deep-reader krafton-ai confluence-deep-reader is a Claude Code skill that traverses Confluence page trees to depth 3 and synthesizes hierarchical documentation. CX Codex +1 1installs

From the same set6

More in the catalog

Project sponsors

Companies that keep this analytics open