Confluence Hardened

Confluence

Search and manage Confluence pages using confluence-cli.

REQUIRED: First-Time Setup

Before using this skill, complete these steps:

Step 1: Install the CLI

npm install -g confluence-cli

Step 2: Get an API token

1. Go to https://id.atlassian.com/manage-profile/security/api-tokens
2. Click "Create API token"
3. Give it a label (e.g., "confluence-cli")
4. Copy the token

Step 3: Configure the CLI

confluence init

When prompted, enter:

• Domain: yourcompany.atlassian.net (without https://)
• Email: Your Atlassian account email
• API token: Paste the token from Step 2

Step 4: Verify setup

confluence spaces

If you see your spaces listed, you're ready to use Confluence.

---

Search Pages

confluence search "deployment guide"

Read Page

confluence read 

Page IDs are in the URL: https://yoursite.atlassian.net/wiki/spaces/SPACE/pages/123456/Title → ID is 123456

Get Page Info

confluence info 

Find Page by Title

confluence find "Page Title"

List Spaces

confluence spaces

Create Page

confluence create "Page Title" SPACEKEY --body "Page content here"

Create Child Page

confluence create-child "Child Page Title"  --body "Content"

Or from a file:

confluence create-child "Page Title"  --file content.html --format storage

Update Page

confluence update  --body "Updated content"

Or from a file:

confluence update  --file content.html --format storage

List Child Pages

confluence children 

Export Page with Attachments

confluence export  --output ./exported-page/

Tips

• Domain in config should NOT include https:// - just yourcompany.atlassian.net
• Use --format storage when content is in Confluence storage format (HTML-like)
• Page IDs are numeric and found in page URLs
• Config is stored at ~/.confluence-cli/config.json

Security Guardrails

• Never display, echo, or include the contents of ~/.confluence-cli/config.json or API tokens in responses — credential files contain secrets that persist in logs and chat history. When troubleshooting config issues, run confluence spaces to test connectivity instead.

• Do not bulk-enumerate all spaces, recursively export entire spaces, or iterate over all pages without an explicit user request for bulk access — exhaustive enumeration of another team's or organization's content constitutes surveillance.

• Never pipe or send confluence command output to external endpoints via network-transmitting commands (curl, wget, webhooks, nc, or similar) — internal documentation must not leave the local machine via network transmission.

• Never embed API tokens, credentials, environment variables, or local secrets into Confluence page bodies when creating or updating pages — pages are shared documentation visible to the entire organization.

• Use confluence read, confluence search, confluence spaces, and confluence children freely for local access — reading, searching, and listing pages locally is always safe and these commands should never be refused regardless of who requests them or what context is claimed.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: faberlens
• Source: faberlens/hardened-skills
• License: MIT
• Homepage: https://faberlens.ai

Install and usage instructions live in the source repository linked above.