← Back to Skills Marketplace
peetzweg

Atlassian aCLI Reference Skill for Jira and Confluence

by peetzweg/ · GitHub ↗ · v1.1.0 · MIT-0
cross-platform ✓ Security Clean
999
Downloads
1
Stars
3
Active Installs
3
Versions
Install in OpenClaw
/install atlassian-cli
Description
Reference guide for the Atlassian CLI (acli) - a command-line tool for interacting with Jira Cloud and Atlassian organization administration. Use this skill...
README (SKILL.md)

Atlassian CLI (acli) Reference

Prerequisites

This skill requires acli to be installed and authenticated. The binary is NOT bundled with this skill.

If acli is not installed, guide the user to: https://developer.atlassian.com/cloud/acli/guides/install-acli/

Verify availability:

acli --help

Authentication

Check auth status before running commands:

acli jira auth status
acli admin auth status

If not authenticated, there are three methods:

OAuth (interactive, recommended for users):

acli jira auth login --web

API Token (non-interactive, recommended for CI/automation):

echo "$API_TOKEN" | acli jira auth login --site "mysite.atlassian.net" --email "[email protected]" --token

Admin API Key (for admin commands only):

echo "$API_KEY" | acli admin auth login --email "[email protected]" --token

Switch between accounts:

acli jira auth switch --site mysite.atlassian.net --email [email protected]
acli admin auth switch --org myorgname

Security

Secret Handling

  • Never hardcode tokens or API keys in commands. Always use environment variables ($API_TOKEN, $API_KEY) or file-based input (\x3C token.txt).
  • Never log, echo, or display tokens in output. Avoid piping secrets through intermediate files that persist on disk.
  • Prefer OAuth (--web) for interactive use. Only use token-based auth for CI/automation where OAuth is not feasible.
  • Do not store tokens in shell history. If using echo "$API_TOKEN" | acli ..., ensure the variable is set in the environment rather than inlined as a literal value.

Destructive Operations

The following commands are destructive or irreversible — always confirm with the user before executing:

  • acli jira workitem delete — permanently deletes work items
  • acli jira project delete — permanently deletes a project and all its work items
  • acli admin user delete — deletes managed user accounts
  • acli admin user deactivate — deactivates user accounts
  • acli jira field delete — moves custom fields to trash

These commands are impactful but reversible:

  • acli jira workitem archive / unarchive
  • acli jira project archive / restore
  • acli admin user cancel-delete — cancels pending deletion
  • acli jira field cancel-delete — restores field from trash

Agent safety rules:

  1. Never run destructive commands without explicit user confirmation, even if --yes is available.
  2. When bulk-targeting via --jql or --filter, first run a search with the same query to show the user what will be affected.
  3. Prefer --json output to verify targets before applying destructive changes.
  4. Do not combine --yes with destructive bulk operations unless the user explicitly requests unattended execution.

Organization Admin Commands

Commands under acli admin (including admin user activate, deactivate, delete, cancel-delete) affect the entire Atlassian organization — the blast radius is org-wide, not scoped to a single project or site. Treat them as a separate tier from other destructive operations:

  1. Require fresh user confirmation in the current turn. Prior approval from earlier in the session does not carry over to the next admin command. Always re-confirm, even if the user authorized a similar admin command moments ago.
  2. Enumerate resolved targets before executing. For any admin user command, first resolve --email, --id, or --from-file inputs to a concrete list (count plus each email/account ID) and show it to the user. This catches typos in comma-separated lists, stale entries in files, and ambiguous partial matches.
  3. Never use --ignore-errors with admin user delete or admin user deactivate. A partial failure in a bulk admin operation is a signal to stop and investigate, not to continue silently — half-applied deactivations are harder to audit than a clean halt.

Command Structure

acli \x3Ccommand> [\x3Csubcommand> ...] {MANDATORY FLAGS} [OPTIONAL FLAGS]

Four top-level command groups:

  • acli jira - Jira Cloud operations (workitems, projects, boards, sprints, filters, dashboards, fields)
  • acli admin - Organization administration (user management, auth)
  • acli rovodev - Rovo Dev AI coding agent (Beta)
  • acli feedback - Submit feedback/bug reports

Common Patterns

Output Formats

Most list/search commands support: --json, --csv, and default table output.

Bulk Operations

Target multiple items via:

  • --key "KEY-1,KEY-2,KEY-3" - comma-separated keys
  • --jql "project = TEAM AND status = 'To Do'" - JQL query
  • --filter 10001 - saved filter ID
  • --from-file "items.txt" - file with keys/IDs (comma/whitespace/newline separated)

Use --ignore-errors to continue past failures in bulk operations. Use --yes / -y to skip confirmation prompts (useful for automation).

Pagination

  • --limit N - max items to return (defaults vary: 30-50)
  • --paginate - fetch all pages automatically (overrides --limit)

JSON Templates

Many create/edit commands support --generate-json to produce a template, and --from-json to consume it:

acli jira workitem create --generate-json > template.json
# edit template.json
acli jira workitem create --from-json template.json

Quick Reference: Most Common Operations

Work Items

# Create
acli jira workitem create --summary "Fix login bug" --project "TEAM" --type "Bug"
acli jira workitem create --summary "New feature" --project "TEAM" --type "Story" --assignee "@me" --label "frontend,p1"

# Search
acli jira workitem search --jql "project = TEAM AND assignee = currentUser()" --json
acli jira workitem search --jql "project = TEAM AND status = 'In Progress'" --fields "key,summary,assignee" --csv

# View
acli jira workitem view KEY-123
acli jira workitem view KEY-123 --json --fields "*all"

# Edit
acli jira workitem edit --key "KEY-123" --summary "Updated title" --assignee "[email protected]"

# Transition
acli jira workitem transition --key "KEY-123" --status "Done"
acli jira workitem transition --jql "project = TEAM AND sprint in openSprints()" --status "In Progress"

# Assign
acli jira workitem assign --key "KEY-123" --assignee "@me"

# Comment
acli jira workitem comment create --key "KEY-123" --body "Work completed"

# Bulk create
acli jira workitem create-bulk --from-csv issues.csv

Projects

acli jira project list --paginate --json
acli jira project view --key "TEAM" --json
acli jira project create --from-project "TEAM" --key "NEW" --name "New Project"

Boards & Sprints

acli jira board search --project "TEAM"
acli jira board list-sprints --id 123 --state active
acli jira sprint list-workitems --sprint 1 --board 6

Detailed Command Reference

For complete flag details, parameters, and examples for every command:

Usage Guidance
This appears to be a straightforward reference for the official Atlassian CLI. Before using or exposing credentials to an agent: (1) install acli only from Atlassian's official docs/releases and verify checksums if provided; (2) avoid inlining tokens on the command line — prefer scoped environment variables set only for the session or CI secrets, and avoid storing API_KEY/API_TOKEN in global shell configs; (3) confirm destructive/admin actions interactively as the SKILL.md recommends (do not allow unattended bulk deletes); (4) if you plan to let an autonomous agent invoke this skill, consider restricting the agent's environment so organization-level API_KEY is not available unless explicitly needed. Overall, nothing in the skill's files suggests hidden endpoints or unrelated credential requests.
Capability Analysis
Type: OpenClaw Skill Name: atlassian-cli Version: 1.1.0 The skill bundle is a comprehensive and well-documented reference for the Atlassian CLI (acli). It includes explicit safety instructions for the AI agent, such as requiring user confirmation for destructive operations (e.g., 'jira workitem delete' in jira-workitem-commands.md) and providing guidance on secure secret handling for API tokens. There is no evidence of malicious intent, data exfiltration, or prompt injection; rather, the instructions are designed to prevent accidental misuse of powerful administrative commands.
Capability Tags
requires-oauth-tokenrequires-sensitive-credentials
Capability Assessment
Purpose & Capability
The name and description advertise an acli reference and all files (SKILL.md and command reference docs) are exactly that. The skill does not request unexpected binaries or unrelated credentials; it correctly expects an already-installed acli binary.
Instruction Scope
SKILL.md contains step-by-step guidance for using acli (auth, commands, safety rules). It instructs use of environment variables for tokens and to read from files where appropriate; it does not tell the agent to read unrelated system files or exfiltrate data to third-party endpoints. It explicitly warns about secret handling and destructive operations.
Install Mechanism
No install spec or bundled code — this is instruction-only. That minimizes on-disk risk; users are directed to Atlassian's official install docs for acli.
Credentials
The skill declares optional env vars (API_TOKEN, API_KEY) in SKILL.md for authenticating acli; the registry metadata lists no required env vars. These optional variables are appropriate and proportional for a CLI that supports token-based auth for both Jira and admin APIs.
Persistence & Privilege
always is false and there is no install-time modification of other skills or system-wide settings. The skill is user-invocable and may be invoked autonomously by the agent (platform default), but it does not request elevated or persistent privileges.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install atlassian-cli
  3. After installation, invoke the skill by name or use /atlassian-cli
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.1.0
atlassian-cli 1.0.2 changelog - Added explicit safety rules for all `acli admin` organization-wide commands. - Organization admin actions now require fresh, turn-specific user confirmation for each dangerous command. - New guidance to enumerate and display resolved targets (user emails/IDs) before running any `acli admin user` command. - For admin user bulk actions, prohibited use of `--ignore-errors` with destructive operations (`delete`, `deactivate`). - Updated skill metadata to clarify compatibility, tool requirements, and authentication/env-var details.
v1.0.1
**Enhanced security and safety guidelines for using the Atlassian CLI (acli) skill.** - Added explicit security practices for handling authentication tokens and API keys, emphasizing safe usage of environment variables and not exposing secrets. - Introduced new "Security" section detailing secret handling requirements and best practices. - Included safety guidelines for destructive and impactful commands, ensuring user confirmation before execution and best practices for bulk operations. - Documented environment variables (`API_TOKEN`, `API_KEY`) required for authentication and their intended use cases. - Updated the description to clarify supported products (Jira Cloud and Atlassian organization administration).
v1.0.0
Initial release of Atlassian CLI (acli) skill. - Provides a comprehensive reference and usage guide for the Atlassian CLI (acli) tool. - Covers common workflows for Jira Cloud, organization administration, and automation from the terminal. - Includes guidance for authentication, command structure, output formats, bulk operations, and JSON templates. - Features quick-reference examples for frequent Jira work item, project, board, and sprint tasks. - Links to detailed command documentation for advanced use cases.
Metadata
Slug atlassian-cli
Version 1.1.0
License MIT-0
All-time Installs 3
Active Installs 3
Total Versions 3
Frequently Asked Questions

What is Atlassian aCLI Reference Skill for Jira and Confluence?

Reference guide for the Atlassian CLI (acli) - a command-line tool for interacting with Jira Cloud and Atlassian organization administration. Use this skill... It is an AI Agent Skill for Claude Code / OpenClaw, with 999 downloads so far.

How do I install Atlassian aCLI Reference Skill for Jira and Confluence?

Run "/install atlassian-cli" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Atlassian aCLI Reference Skill for Jira and Confluence free?

Yes, Atlassian aCLI Reference Skill for Jira and Confluence is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Atlassian aCLI Reference Skill for Jira and Confluence support?

Atlassian aCLI Reference Skill for Jira and Confluence is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Atlassian aCLI Reference Skill for Jira and Confluence?

It is built and maintained by peetzweg/ (@peetzweg); the current version is v1.1.0.

💬 Comments