From jira
Generate a condensed project-wide product roadmap from Jira version data. Produces a multi-version overview with theme summaries, highlights, and links to detailed release notes pages. Use when the user wants a product roadmap, project overview, or version overview. Triggers on any request mentioning roadmap, product roadmap, project overview, or version overview.
How this skill is triggered — by the user, by Claude, or both
Slash command
/jira:jira-roadmapThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Generate a professional, client-facing product roadmap from Jira version data. Produces a condensed project-wide overview with version summaries, highlights, and links to individual release notes pages. Uses the `jira-fetch` script to pull issue data via REST API — no MCP overhead, no subagents, no token waste.
Generate a professional, client-facing product roadmap from Jira version data. Produces a condensed project-wide overview with version summaries, highlights, and links to individual release notes pages. Uses the jira-fetch script to pull issue data via REST API — no MCP overhead, no subagents, no token waste.
AskUserQuestionjira-fetch to discover versions, then fetch per-version dataBefore anything else, use a single AskUserQuestion call with two questions:
If Confluence output is selected, immediately follow up with one more AskUserQuestion (header: "Confluence URL") asking for the target Confluence page URL (the full URL, e.g. https://mycompany.atlassian.net/wiki/spaces/PROJ/pages/123456).
If $ARGUMENTS contains a project name or key (e.g., SF, ShopFlow), extract it for use in Step 2 to skip project discovery.
Use the selected language for the entire document. Translate section headers according to the translations in the respective format reference file.
Look for a JIRA configuration block in the project's CLAUDE.md:
## JIRA
- Domain: mycompany.atlassian.net
- Project key: PROJ
If found, use that domain and project key. If not found, ask via AskUserQuestion (header: "JIRA Configuration"):
Store the resolved domain, projectKey, and base URL (https://{domain}) for the rest of the session. The base URL is needed for clickable task links in the output.
This step discovers available versions, lets the user select which ones to include, and then fetches full issue data per version. Uses the jira-fetch script for all data retrieval — no MCP tools needed.
Find the fetch script via Glob:
pattern: **/jira-fetch/scripts/fetch-issues.mjs
Run jira-fetch to get all issues with version assignments:
node "${SCRIPT_PATH}" \
--domain "${DOMAIN}" \
--jql "project = ${PROJECT_KEY} AND fixVersion IS NOT EMPTY ORDER BY fixVersion DESC" \
--output "/tmp/jira-roadmap-versions-$(date +%Y%m%d-%H%M%S).json"
Read the JSON output and extract unique version names from the fixVersions field across all issues. Present discovered versions to the user via AskUserQuestion (header: "Versions") to confirm which versions to include. Allow the user to deselect versions they want to exclude.
If the script fails, show the error and stop. Common issues: missing JIRA_EMAIL or JIRA_API_TOKEN env vars.
The discovery data already contains full issue details. Filter it by each selected version — for each version, extract the subset of issues where fixVersions includes that version.
For each selected version, compute aggregate stats:
Descriptions are available directly from the JSON data for the top 5 issues per version (by priority: Epics and Stories first, then remaining by priority descending). Having descriptions available enables accurate client-focused version summaries without subagent extraction.
[Project] Release Notes — Version {version} pattern using searchConfluenceUsingCql[Project] Roadmap — Version {version} for backward compatibilitydocs/release-notes/release-notes-{version}.mddocs/roadmaps/roadmap-{version}.md for backward compatibilityAskUserQuestion (header: "File paths") if not foundFor each version, produce:
Use descriptions from the JSON data for the top issues to generate theme names, summaries, and highlights — these provide the detail needed for accurate business-value transformation. For issues not among the top 5, use the summary field.
Frame every summary and highlight from the client's perspective — what problem does this solve for them? How does it make their work easier or their business better? Every highlight should answer: "What does the client gain from this?"
Apply business-value transformation rules from the writing guidelines in references/format.md (no jargon, active voice, client-benefit focus).
Reference format: references/format.md
Assemble the document following the structure in references/format.md:
# [Project] — Product Roadmap
## Overview
[2-3 sentences about overall product direction and strategy]
---
## Version X.Y.Z — [Theme Name]
**Status:** In Progress | **Target:** Q1 2026
[2-3 sentence summary]
**Highlights:**
- Highlight 1
- Highlight 2
- Highlight 3
[→ Release notes](link)
Version ordering: most recent/current version first, then upcoming, then past releases.
Full example: references/example.md
Present the complete roadmap inside a clearly marked block and use AskUserQuestion (header: "Review"):
If the user asks for adjustments, apply changes and present the updated draft again. Repeat until the user confirms.
Present the final roadmap in a code block and suggest saving to a file:
Your roadmap is ready. Want me to save it to
docs/roadmaps/roadmap-main.md?
If the user confirms, write the file.
Confluence publishing requires a cloudId. Resolve it now by calling getAccessibleAtlassianResources MCP tool — this is the only MCP call in this skill, and only when Confluence output was selected.
Use the available Confluence MCP tools to publish:
[Project] — Product Roadmap in the project's Confluence spaceAdd links to existing release notes pages within each version section. If no Confluence MCP tools are found, fall back to markdown output and inform the user.
After publishing, provide the page URL to the user.
npx claudepluginhub mrstroz/claude-code-plugins --plugin jiraGenerates project status reports from Jira issues using JQL, categorizes by status/priority, analyzes blockers/progress, formats for executives/teams, and publishes to Confluence.
Provides templates for Confluence docs like TDD, ADR, API, runbooks, release notes. Covers CQL queries and Jira linking for engineering teams managing technical documentation.
Builds a product roadmap from strategy, objectives, and connector data. Supports Shape Up, Scrum, and NOW/NEXT/LATER methodologies.