Project Management

This skill enables an AI agent to serve as a project management co-pilot for software teams. It covers the full lifecycle: breaking high-level goals into actionable tasks with acceptance criteria, estimating effort in story points, organizing work into sprints or Kanban boards, tracking progress with burndown data, identifying blockers and risks, and generating stakeholder-ready status reports. The agent integrates with Jira, Linear, GitHub Issues, and Trello so that all changes are reflected in the team's actual project tracker.

Workflow

1. Gather Project Context Collect the project name, team roster (with roles and capacity), high-level goals, key deadlines, and the preferred methodology (Scrum, Kanban, or a hybrid). If a project tracker is already in use, connect to its API and import the current backlog, labels, and sprint configuration so the agent starts from the existing state rather than a blank slate.

2. Decompose Work into Tasks Break each high-level goal into epics, then into user stories, and finally into sub-tasks small enough to complete in one to two days. Write each user story in the standard format: "As a [role], I want [capability] so that [benefit]." Attach acceptance criteria as a checklist on every story. Identify dependencies between tasks and flag any that form a critical path. Assign story-point estimates using the Fibonacci scale (1, 2, 3, 5, 8, 13) based on complexity, uncertainty, and effort.

3. Plan Sprints or Populate the Board For Scrum teams, create a sprint with a start date, end date, and sprint goal. Pull stories from the top of the prioritized backlog until the team's velocity (historical average story points per sprint) is reached. For Kanban teams, enforce work-in-progress (WIP) limits per column and ensure the board columns match the team's workflow (e.g., Backlog → In Progress → Review → Done). Push the plan to the connected project tracker via its API.

4. Track Progress and Surface Blockers Poll the project tracker on a configurable interval (default: daily). Calculate burndown and cumulative-flow metrics. Detect when a task has been "In Progress" longer than its estimate and flag it as at risk. Identify dependency chains where a blocked upstream task is stalling downstream work. Post a daily digest to the team's Slack or Teams channel summarizing completed items, in-progress items, and blockers.

5. Manage Risks and Dependencies