# Jira Buddy – Overview & Vision ## What Are We Trying to Achieve with Jira Buddy? Jira Buddy is an intelligent assistant designed to help teams **understand, manage, and act on Jira data using natural language**. Its goal is to remove the day-to-day friction of working inside Jira and allow users to focus on delivery rather than administration. Instead of navigating complex Jira screens, workflows, and filters, users can simply ask Jira Buddy questions or give it commands. ## The Problem Jira Buddy Solves Jira is powerful, but it is often: - Time-consuming to operate - Difficult to query meaningfully - Fragmented across projects and teams - Heavy on manual updates and reporting Jira Buddy exists to **simplify Jira usage without reducing its power**. ## What Jira Buddy Does Jira Buddy acts as: ### 1. A Jira Interpreter It converts raw Jira data into clear, human-readable insights: - Sprint health - Blockers - Ownership clarity - Change tracking ### 2. A Jira Operator It performs Jira actions safely on behalf of users: - Create and update issues - Transition statuses - Assign owners - Add comments and metadata ### 3. A Context-Aware Assistant It understands: - Project structures - Team responsibilities - Workflow rules - Historical context This ensures actions are accurate and compliant with Jira constraints. ### 4. A Decision Support Tool It highlights risks, bottlenecks, and inefficiencies early so teams can act proactively. ## How to Read This Document This document is structured into multiple sections (sheets), each representing a different part of the Jira Buddy system: - Product intent and scope - Technical integrations - User workflows and use cases - Execution and implementation tasks Each section contributes to the overall goal of building a reliable, intelligent Jira assistant. --- # Layer Scope Overview This section documents the **Layer Scope Overview** sheet. It includes details across the following columns: Layer, Key Question, Primary Goal, In-Scope Capabilities, Out of Scope, Success Indicators. | Layer | Key Question | Primary Goal | In-Scope Capabilities | Out of Scope | Success Indicators | |:-------------|:---------------------------|:-------------------------|:------------------------------------------------------|:----------------------------------------|:-------------------------------| | Descriptive | What happened? | Conversational reporting | NL→JQL, validation, counts, lists, trends, drill-down | Root cause, prediction, recommendations | High trust, reduced Jira usage | | Proactive | What should I worry about? | Early warnings | Baselines, alerts, trend detection | Deep diagnosis, optimization | Issues caught early | | Diagnostic | Why did it happen? | Root cause explanation | Segmentation, comparisons, driver analysis | Predictions, automation | Faster retros | | Predictive | What will happen? | Risk forecasting | Sprint risk, SLA risk, backlog forecast | Auto decisions | Fewer surprises | | Prescriptive | What should we do? | Decision support | Recommendations, what-if analysis | Auto execution | Advice adoption | | Autonomous | Do it continuously | Closed-loop automation | Auto-assign, escalate, policy actions | Uncontrolled actions | Manual effort reduction | # Development Tasks This section documents the **Development Tasks** sheet. It includes details across the following columns: Layer, Epic, Dev Task, Description, Dependencies, Complexity (L/M/H), Owner Role. | Layer | Epic | Dev Task | Description | Dependencies | Complexity (L/M/H) | Owner Role | |:-------------|:----------------|:------------------------|:-----------------------------------------------------|:------------------------|:---------------------|:-------------| | Descriptive | NL → JQL | Intent classification | Detect user intent like count, list, trend | Chat input pipeline | M | Backend / AI | | Descriptive | NL → JQL | Entity extraction | Extract project, assignee, status, dates | Intent classification | M | Backend / AI | | Descriptive | Query Engine | JQL builder | Generate valid JQL from structured intent | Entity extraction | M | Backend | | Descriptive | Query Engine | JQL validation & repair | Validate JQL and auto-correct common errors | JQL builder | H | Backend | | Descriptive | Results | Result summarization | Convert raw Jira results into NL summary | Query execution | M | AI | | Proactive | Monitoring | Scheduled query runner | Run saved queries on schedule | Query engine | M | Backend | | Proactive | Insights | Baseline computation | Compute rolling averages and comparisons | Historical data storage | M | Backend | | Proactive | Alerts | Alert triggering logic | Trigger alerts based on thresholds/trends | Baseline computation | M | Backend | | Diagnostic | Analysis | Segmentation engine | Group issues by dimension (assignee, priority, etc.) | Query engine | M | Backend | | Diagnostic | Analysis | Contribution analysis | Identify drivers contributing to changes | Segmentation engine | H | AI / Backend | | Diagnostic | Explainability | Root cause narration | Generate NL explanation with evidence | Contribution analysis | M | AI | | Predictive | Modeling | Feature engineering | Prepare features like velocity, aging | Historical data | H | Data / ML | | Predictive | Modeling | Risk scoring model | Predict sprint/SLA risk probabilities | Feature engineering | H | ML | | Predictive | Serving | Prediction API | Serve predictions to chat & insights | Risk model | M | Backend | | Prescriptive | Simulation | What-if engine | Simulate outcomes for different actions | Predictive outputs | H | Backend / ML | | Prescriptive | Recommendations | Action ranking | Rank actions by impact & feasibility | What-if engine | M | AI | | Autonomous | Automation | Policy engine | Define rules and guardrails for actions | Prescriptive layer | H | Backend | | Autonomous | Automation | Approval workflow | Human-in-the-loop approvals | Policy engine | M | Backend | | Autonomous | Safety | Audit & rollback | Log actions and enable rollback | Automation executor | M | Backend |