Jira Team Management Dashboard

A Streamlit application that fetches Jira tickets via JQL and renders delivery analytics dashboards for team leads and engineering managers.

Features

• Delayed Tasks Trend — weekly count of tickets past their ETA per user
• Total Delay per User — cumulative delay days per user per week (bar or area chart)
• Delivery Velocity — tickets completed per user per week
• Status Distribution — ticket breakdown by workflow status (pie + bar + per-user stacked bar)
• Cycle Time Analysis — median/mean creation-to-resolution time per user with weekly trend
• Priority Breakdown — tickets by priority over time and per user
• Unassigned Tasks — tickets with no assignee; supports assigning a team member in-app
• Workload Overview — active tasks and story point share per person
• Team Members — directory of configured Jira users with names and emails

Requirements

• Python 3.10+
• A Jira Cloud instance with API access
• An Atlassian API token — generate one at https://id.atlassian.com/manage-profile/security/api-tokens

Setup

1. Install dependencies

pip install -r requirements.txt

2. Configure credentials

cp .env.example .env

Edit .env and fill in:

Variable	Description
JIRA_BASE_URL	Your Atlassian instance URL, e.g. https://your-domain.atlassian.net
JIRA_EMAIL	Your Atlassian account email
JIRA_API_TOKEN	API token from Atlassian (the app encodes it automatically)
JIRA_USERS	Comma-separated Jira account IDs to track, e.g. accountId1,accountId2
JQL_FILTER	Default JQL shown in the sidebar, e.g. type = "Story" ORDER BY created DESC
MAX_RESULTS	Tickets per page — pagination is automatic (default: 100)
ETA_FIELD	Jira field used as the deadline for delay calculation (default: duedate)
PORT	Streamlit server port (default: 8501)

To find a user's account ID: open any ticket in Jira, click the assignee avatar — the profile URL ends with the account ID.

3. Run the app

./run.sh

The script reads PORT from .env and starts Streamlit. Open http://localhost:8501 in your browser.

Project Structure

app.py                      # Sidebar nav, data fetch, page routing
config.py                   # Loads .env into constants
jira_client.py              # Paginated Jira POST /search client (cached 5 min)
logger.py                   # Application logger → application_output.log
utils/
  data_processing.py        # Shared helpers: delay calc, week bucketing, cycle days
  pdf_export.py             # PDF download button for charts
dashboards/
  delayed_tasks.py          # Delayed Tasks Trend
  total_delay.py            # Total Delay per User
  delivery_velocity.py      # Delivery Velocity
  status_distribution.py    # Status Distribution
  cycle_time.py             # Cycle Time Analysis
  priority_breakdown.py     # Priority Breakdown
  unassigned_tasks.py       # Unassigned Tasks
  workload.py               # Workload Overview
  users.py                  # Team Members
tests/
  conftest.py               # Shared fixtures and sample data factory
  test_data_processing.py   # Unit tests for utils/data_processing.py
  test_dashboard_logic.py   # Unit tests for dashboard business logic
  test_ui.py                # End-to-end Playwright UI tests

Running Tests

Unit tests (no app needed)

pytest tests/test_data_processing.py tests/test_dashboard_logic.py -v

UI tests (requires running app)

./run.sh &
pytest tests/test_ui.py -v --browser chromium

UI tests use the MCP Playwright server in headed mode. See tests/TEST_CASES.md for a full description of every test case.

How It Works

Sidebar JQL input
  → jira_client.fetch_tickets(jql)    # paginated POST /rest/api/3/search
  → data_processing.to_df(tickets)    # normalise + parse date columns
  → st.session_state.df               # shared across all dashboards
  → dashboards/<name>.render(df)      # each dashboard filters/groups independently

Data is cached for 5 minutes. Click Fetch Data again to force a refresh.

Security

• .env is gitignored and never read directly by any dashboard or test.
• All credentials are accessed through config.py constants only.
• A pre-tool hook (check_secrets.py) actively blocks any attempt to read .env files directly.
• The app encodes JIRA_EMAIL:JIRA_API_TOKEN to Base64 automatically — no manual step needed.
• No Jira write operations are performed except the optional "Update in Jira" action on the Unassigned Tasks dashboard (requires explicit user interaction).