How do I install the Jira Board skill?

Run npx skillmds@latest add arfar-x/jira-board in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Jira Board safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Jira Board?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Jira Board free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Jira Board?

arfar-x (@arfar-x) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Jira Board by arfar-x — Agent Skill

Name: Jira Board
Author: arfar-x

Jira Board

What's happening on my board right now -- the active sprint if this is a Scrum project, or the column/status breakdown if it's Kanban. Auto-detects which one applies, so you don't need to know in advance. Use for "what's going on with my board", "what's the current sprint", "how's the board looking" when you're not sure which type of project this is.

arfar-x Updated Sep 21, 2026

File contents

Jira: Board

Read-only. This skill routes to whichever of sprint/kanban_status fits -- run from this skill's directory:

python3 ../jira/scripts/jira_tool.py sprint [--project PAY] [--board_id 42]

(First-time setup, once per environment: pip install -r ../jira/requirements.txt.)

Board type comes from memory, not a guess

A project's board type (Scrum, with sprints, vs. Kanban, without) essentially never changes -- unlike the sprint/column data itself, which you always need fresh, every call. This skill never re-derives the type by speculatively calling one command to see what it says; it looks the type up first:

Type already known (your runtime's persistent memory, jira-project-context, or an earlier call this conversation already told you): call the matching command directly and report its live result -- you're never skipping the actual data fetch, only the question of which command to run.
```
# Scrum:
python3 ../jira/scripts/jira_tool.py sprint [--project PAY] [--board_id 42]
# Kanban:
python3 ../jira/scripts/jira_tool.py kanban_status [--project PAY] [--board_id 42]
```
Type not known yet for this project: run sprint -- this is the first real request for this project's board, not a guess. Its result answers the question either way: an active sprint (Scrum), or a "note" saying the board is kanban (report via kanban_status instead, using the same --project/--board_id). The type it reveals is a side effect of a request you needed to make anyway, not an extra probe.
If "board" is null (no board found for the resolved scope), relay the "note" naming the project that was searched -- don't guess or invent a board.

Remember the type the instant you know it -- in this same turn, before the user has to ask. Save it the same way jira-project-context asks you to save a project's statuses/team/labels (see ../jira/README.md's "Agent memory" section for the full catalog) -- this memory is self-learning, and saving is not a follow-up step triggered by the user checking whether you'll remember; it happens as part of reporting the result itself. From then on, every future call (from this skill, jira-sprint, or jira-kanban-status) goes straight to step 1 for this project; step 2 should only happen once per project, ever.

Already know which type of project this is? jira-sprint/ jira-kanban-status call the same underlying commands directly, without this skill's type lookup.

Never invent data not present in either tool's JSON. If a result contains "error", tell the user what went wrong in plain language instead of retrying silently or fabricating a result.

See ../jira/README.md for architecture details and the full environment-variable table.

---name: jira-boarddescription: What's happening on my board right now -- the active sprint if this is a Scrum project, or the column/status breakdown if it's Kanban. Auto-detects which one applies, so you don't need to know in advance. Use for "what's going on with my board", "what's the current sprint", "how's the board looking" when you're not sure which type of project this is.---# Jira: BoardRead-only. This skill routes to whichever of `sprint`/`kanban_status` fits-- run from this skill's directory:```bashpython3 ../jira/scripts/jira_tool.py sprint [--project PAY] [--board_id 42]```(First-time setup, once per environment: `pip install -r ../jira/requirements.txt`.)## Board type comes from memory, not a guessA project's **board type** (Scrum, with sprints, vs. Kanban, without)essentially never changes -- unlike the sprint/column data itself, whichyou always need fresh, every call. This skill never re-derives the typeby speculatively calling one command to see what it says; it looks thetype up first:1. **Type already known** (your runtime's persistent memory, `jira-project-context`, or an earlier call this conversation already told you): call the matching command directly and report its live result -- you're never skipping the actual data fetch, only the question of which command to run. ```bash # Scrum: python3 ../jira/scripts/jira_tool.py sprint [--project PAY] [--board_id 42] # Kanban: python3 ../jira/scripts/jira_tool.py kanban_status [--project PAY] [--board_id 42] ```2. **Type not known yet for this project:** run `sprint` -- this is the first real request for this project's board, not a guess. Its result answers the question either way: an active sprint (Scrum), or a `"note"` saying the board is kanban (report via `kanban_status` instead, using the same `--project`/`--board_id`). The type it reveals is a side effect of a request you needed to make anyway, not an extra probe.3. If `"board"` is null (no board found for the resolved scope), relay the `"note"` naming the project that was searched -- don't guess or invent a board.**Remember the type the instant you know it -- in this same turn, beforethe user has to ask.** Save it the same way `jira-project-context` asksyou to save a project's statuses/team/labels (see `../jira/README.md`'s"Agent memory" section for the full catalog) -- this memory isself-learning, and saving is not a follow-up step triggered by the userchecking whether you'll remember; it happens as part of reporting theresult itself. From then on, every future call (from this skill,`jira-sprint`, or `jira-kanban-status`) goes straight to step 1 for thisproject; step 2 should only happen once per project, ever.Already know which type of project this is? `jira-sprint`/`jira-kanban-status` call the same underlying commands directly, withoutthis skill's type lookup.Never invent data not present in either tool's JSON. If a resultcontains `"error"`, tell the user what went wrong in plain languageinstead of retrying silently or fabricating a result.See `../jira/README.md` for architecture details and the fullenvironment-variable table.

Frequently asked questions