Menú
A concise reference for using the Atlassian CLI (acli) to manage Jira issues efficiently. Use this skill to quickly search, edit, and create Jira work items across projects or boards.
Instalar con Codex o Claude Copia este prompt, pégalo en Codex, Claude u otro asistente, y deja que revise la página de la skill y la instale por ti.
Basado en la clasificación ocupacional SOC
Create and update a structured Design Review / Software Requirements Specification (DR/SRS). Use whenever a user asks for a Design Review, DR, SRS, solution design, technical requirements document, architecture review, or to turn an accepted RFC into the operational design. Write Markdown first, preserve stable IDs, expose gaps without inventing decisions, invoke uc-engraver when detailed Use Cases are requested, and hand confirmed Confluence publication to the confluence-librarian skill.
Project approved DR/SRS and Use Case contracts into a reviewable Jira backlog. Use whenever a user asks to create, decompose, refine, synchronize, or update Jira Epics, User Stories, or technical Tasks from a PRD, DR/SRS, Use Case, JTBD, acceptance check, RFC, or requirements document. Propose multiple cohesive Epics that fit a few sprints, write actor-facing Stories sized for one two-week sprint, separate enabling work into Tasks, preserve source traceability, and require explicit confirmation before Jira mutations.
Create and update structured Use Case child documents for a parent Design Review / Software Requirements Specification (DR/SRS). Use whenever a user asks to define, write, refine, synchronize, or publish a Use Case, UC, scenario, acceptance checks, or detailed behavior from a PRD, DR/SRS, JTBD, Service Blueprint, Figma, RFC, contract, or task intake. Preserve stable UC/AC IDs, reuse actor names exactly from the linked PRD, expose gaps without inventing behavior, update and verify the parent DR/SRS catalog, and hand confirmed Confluence publication to confluence-librarian.
Publish and update prepared pages and document artifacts in Confluence without losing structure or traceability. Use whenever a user asks to publish, create, update, translate, or synchronize Markdown, HTML, or another readable document to Confluence, including hierarchical child pages under a parent catalog. Require explicit confirmation before irreversible publication and preserve headings, tables, links, statuses, IDs, and intentional gaps.
Write and update a structured Product Requirements Document (PRD) from gathered product material. Use when a product manager wants to turn discovery notes, meeting summaries, strategy input, or a rough product document into a Confluence-compatible PRD covering needs, actors, JTBD, goals, metrics, guardrails, dependencies, constraints, feasibility red flags, design discovery, open questions, and support readiness. Preserve stable English IDs, manage explicit draft-to-review readiness, and offer optional Confluence publication only after writing.
Build or extend backend tests for Node.js/TypeScript services with real runtimes, Testcontainers dependencies, optional .env.test cloud services, and integration and/or record-replay coverage. Use for integration tests, live contracts, freeze-before-refactor, VCR/golden-master, cassette record/verify, Azure Functions harnesses, replacing mock-heavy tests, or adding scenarios to an existing backend-test harness. Always use when choosing between integration and record-replay.
| name | jira-cli |
| description | A concise reference for using the Atlassian CLI (acli) to manage Jira issues efficiently. Use this skill to quickly search, edit, and create Jira work items across projects or boards. |
This skill provides a quick reference for common commands and operations related to Jira work items.
Before using the Atlassian CLI (acli), ensure you have it installed and that the user is authenticated and has the necessary permissions to perform the desired operations in Jira.
Install instructions, in case acli binary isn't yet installed, can be found in the Atlassian CLI documentation.
The term workitem is used here to refer to any Jira issue, including tasks, bugs, stories, etc. Any reference to "work item" can be substituted with "issue" or the specific issue type as needed and vice versa.
Start by confirming that the Atlassian CLI is installed and available in the system PATH by running:
acli --version
# List all acli commands
acli --help
# Jira-specific commands
acli jira --help
Run
acli jira project view --key <PROJECT-KEY>
to verify current authentication: acli auth status does not work reliably for this purpose.
If not authenticated, point user to Atlassian CLI Getting Started Guide.
# Search with JQL query (returns JSON)
acli jira workitem search --jql 'project = CES AND creator = currentUser()' --json
# Search with JQL, limit and specific fields
acli jira workitem search --jql 'project = CES AND status IN (Backlog, "In Progress") AND issuetype IN (Task, Bug) AND description is EMPTY' --fields 'key,summary,status,issuetype' --limit 100 --json
# Count results
acli jira workitem search --jql 'project = CES' --count
# Get details by key
acli jira workitem view CES-123 --json
# Update a single work item by key
acli jira workitem edit --key CES-1796 --description "Your description" --yes
# Update multiple work items with JQL
acli jira workitem edit --jql 'project = CES' --assignee "user@example.com" --yes
# Update from JSON file
acli jira workitem edit --from-json "workitem.json"
# Generate JSON template for editing
acli jira workitem edit --generate-json
# Create a new work item
acli jira workitem create --project CES --type Task --summary "Title" --description "Description"
# Bulk create from JSON
acli jira workitem create-bulk --from-json "items.json"
--key: Specify work item key(s) (e.g., "CES-1796" or "CES-1796,CES-1797")--jql: Use JQL query to target multiple items--json: Output results in JSON format--csv: Output results in CSV format--fields: Specify comma-separated fields to display (default: issuetype,key,assignee,priority,status,summary)--description: Set/update description (supports plain text or ADF format)--yes: Confirm action without prompting (useful for scripts)--limit: Maximum number of items to fetch--paginate: Fetch all results by paginating through results--json output for parsing with jq (powerful for filtering and transforming results)--yes flag when scripting to avoid interactive prompts