agentleFS
Sign inStart free
dbhq-ukdbhq-uk/atlassian-skill/skills/atlassian/SKILL.md

atlassian

Skill0 starsChanged 49 days agoMITView on GitHub

Jira Cloud and Confluence Cloud over the REST API, with an API token and no MCP server. Creates, reads and comments on Jira issues and moves one through its workflow, searches, reads, creates and updates Confluence pages, and publishes a repository's markdown file to a Confluence page. Never deletes. Use whenever the user mentions Jira, Confluence or "the wiki", writes JQL or CQL, wants a ticket raised, read, searched, commented on or moved, wants a wiki page found, written or updated, or wants repository docs published or synced to Confluence.

5 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions5 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 5

    1. sanjay3290/ai-skills★ 452same name
    2. nordz0r/skills★ 8same name
    3. razbakov/skills★ 4same name
    4. andeeplus/aplus★ 2same name
    5. sanjay3290/atlassian-skill★ 0same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repositoryOne other file it gives its agents.

The file

We split this skill into the rules it gives an agent

6 rules. 0 are common, 3 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 5 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read the matching file before you run anything.

    line 14
  2. Very specific:

    Every command in those files starts with the CLAUDE_SKILL_DIR placeholder.

    line 22
  3. Distinctive:

    Before you write a Jira description, a Confluence page or a file you will publish, read:

    line 26
  4. Very specific:

    Check before you write: Never invent a Jira project key, issue type or field name - read them with jira-meta.sh.

    line 50
  5. Distinctive:

    Never put a credential, a token or personal data into an issue or a page.

    line 56
  6. Very specific:

    When a script fails, read the Cause: and Fix: lines it prints.

    line 58

“Common” is measured: kept by at least 6 in 10 of the 5 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

atlassian

Jira Cloud and Confluence Cloud through their REST APIs, on one API token. No MCP server.

This skill creates, reads and updates. It never deletes - not an issue, a page, an attachment or a space. It has no bulk transition and no project or space administration. Anything destructive is a human job in the Atlassian UI.

Read the reference for the task

Read the matching file before you run anything. Each one holds the commands, the checks and the error messages for its task.

TaskRead
Create, read, search or comment on Jira issues, or move one to another status${CLAUDE_SKILL_DIR}/references/jira.md
Search, read, create, edit or update a Confluence page${CLAUDE_SKILL_DIR}/references/confluence.md
Publish a markdown file to a Confluence page${CLAUDE_SKILL_DIR}/references/publish.md

Every command in those files starts with the CLAUDE_SKILL_DIR placeholder. Read it as this skill's own directory: ${CLAUDE_SKILL_DIR}. Every script the skill runs is inside it.

Before you write anything

Before you write a Jira description, a Confluence page or a file you will publish, read:

  1. ${CLAUDE_SKILL_DIR}/references/house-style.md - the conventions
  2. ${CLAUDE_SKILL_DIR}/references/html-patterns.md - the HTML+ patterns
  3. ~/.dbhq/atlassian/house-style.md, if it exists - the user's own conventions, which win over the shipped ones

Then work through ${CLAUDE_SKILL_DIR}/references/checklist.md against the body before you send it. Jira issue text also follows ${CLAUDE_SKILL_DIR}/references/ste.md.

Credentials

One credential serves both products: ~/.dbhq/atlassian/config.json, mode 600, with the site URL, the account email and an API token.

If that file does not exist, setup has not run. Do not run setup yourself. It asks for the token with the input hidden, so it needs the user at a terminal, and the token must not pass through this conversation. Give the user this command, with the full path written out, and ask them to run it in their own terminal:

${CLAUDE_SKILL_DIR}/scripts/atlassian-setup.sh

Every script prints the same full path when it finds no credential. Setup asks for the site URL, the email and an API token from https://id.atlassian.com/manage-profile/security/api-tokens. It verifies Jira access before it saves anything, then checks Confluence access (a warning, not a blocker). The token never reaches a command line: curl reads it from a 0600 config file.

The scripts need jq, curl, column and Python 3 (standard library only).

The rules

  1. Check before you write. Never invent a Jira project key, issue type or field name - read them with jira-meta.sh. Never invent a Confluence media id or collection.
  2. Confirm with the user before you create, comment or move an issue. An issue, a comment or a page is visible to the team the moment it exists, and this skill cannot delete it.
  3. Dry-run first for jira-issues.sh bulk and for publish.sh.
  4. A Confluence update or edit needs --base-version - the version your edit started from. It refuses if the page has moved on. To change part of a page, use edit, which leaves the rest untouched.
  5. Bodies are HTML+, converted to ADF and checked locally before anything is sent. Never markdown on the wire, never storage format.
  6. There is no --force. A refusal needs a human decision, not a flag.
  7. Never put a credential, a token or personal data into an issue or a page.

When a script fails, read the Cause: and Fix: lines it prints.

When not to use it

  • GitHub issues or pull requests. Use gh.
  • Trello boards and cards.
  • Jira or Confluence Server or Data Center. This skill speaks the Cloud REST APIs only.
  • A session that already has the Atlassian MCP server connected, unless the user asks for this skill by name.
  • Deleting anything, or moving many issues at once. That stays in the Atlassian UI.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free