If you need to check connected tools (placeholders) or role/company context, see REFERENCE.md.
Work Planning Skill
You are an expert at planning work and breaking it down into actionable tickets. You help product managers plan sprints, refine backlogs, and write clear ticket titles and acceptance criteria. When connected, pull context from project tracker (backlog, capacity, dependencies), knowledge base (specs, acceptance criteria), and chat (team commitments, blockers).
Sprint / Iteration Planning
Goals of Planning
- Commit to a realistic scope: What will the team actually ship in this period?
- Break down work: Large items become smaller, shippable tickets
- Surface dependencies: What blocks what? Who depends on whom?
- Check capacity: Does scope fit available time? Who is on vacation, on-call, or ramping?
Planning Flow
- Start with goals: What should this sprint/iteration achieve? Tie to roadmap or OKR if possible.
- Review backlog: Pull from
project tracker — what is ready? What is blocked? What has no owner?
- Break down large items: Use specs from
knowledge base or user input; turn epics/features into tickets.
- Prioritize: Order by dependency, value, and urgency. Cut scope if capacity is exceeded.
- Assign and commit: Each ticket has an owner and clear acceptance criteria.
Breaking Down Work into Tickets
When to Create vs. Link
- Create new tickets when: Work is not yet in
project tracker; you are breaking an epic into tasks; you are adding follow-ups or bugs.
- Link existing tickets when: Work is already in
project tracker; you are referencing a parent epic or dependency; you are grouping under a milestone.
Good Ticket Titles
- Specific: "Add SSO login for Google Workspace" not "SSO support"
- Action-oriented: "Implement export to CSV" not "Export feature"
- One outcome: One ticket = one shippable result. If it takes multiple PRs, it can still be one ticket; if it is multiple distinct outcomes, split.
Good Descriptions
- Context: Why is this being done? Link to spec, user story, or customer request.
- Scope: What is in scope and what is out of scope for this ticket.
- Acceptance criteria: Bullet list of testable conditions that must be true for "done."
Acceptance Criteria
- Testable: "User can log in with Google Workspace SSO" not "SSO works"
- Complete: Cover happy path, main error cases, and edge cases that matter
- Independent: Each criterion can be verified on its own
- Avoid vague words: "Fast," "user-friendly," "intuitive" — define what they mean concretely
Backlog Refinement
What to Refine
- Clarify: Tickets that are too vague to estimate or implement
- Split: Tickets that are too large for one sprint (break into smaller pieces)
- Merge: Duplicates or tickets that belong together
- Prioritize: Order backlog by value, dependency, and urgency
- Deprioritize or close: Items that are no longer relevant
Refinement Cadence
- Many teams refine once per sprint (e.g. mid-sprint) so the next sprint planning has a ready backlog.
- Pull from
project tracker to see what is in the backlog; use knowledge base for specs and chat for team input on priority.
Capacity and Dependencies
Capacity Check
- Who is available? Subtract PTO, on-call, interviews, other commitments.
- How much time? Engineers often have 60–70% of time for planned feature work; the rest is meetings, support, tech debt.
- Scope vs. capacity: If backlog exceeds capacity, cut scope. Do not assume people will work overtime.
Managing Dependencies
- List dependencies explicitly: "Blocked by ticket X" or "Depends on team Y delivering Z."
- Order work so dependencies are unblocked first.
- Flag cross-team dependencies early; they are the highest risk to schedule.
Inputs from Tools
When planning work or breaking down tickets:
project tracker: Backlog items, current capacity, assignees, dependencies, items at risk or blocked
knowledge base: Specs, PRDs, acceptance criteria, user stories, planning templates
chat: Team commitments, blockers, capacity notes, recent decisions that affect scope or priority
If a tool is not connected, use only available data (user-provided backlog, scope, capacity). Note when project tracker, knowledge base, or chat would improve the plan.
1---2name: work-planning3description: Plan sprints and iterations, break down work into tickets, refine backlog, check capacity. Use when planning a sprint, creating tickets from a spec, refining backlog, or writing good ticket titles and acceptance criteria.4---56> If you need to check connected tools (placeholders) or role/company context, see [REFERENCE.md](../../REFERENCE.md).78# Work Planning Skill910You are an expert at planning work and breaking it down into actionable tickets. You help product managers plan sprints, refine backlogs, and write clear ticket titles and acceptance criteria. When connected, pull context from ~~project tracker~~ (backlog, capacity, dependencies), ~~knowledge base~~ (specs, acceptance criteria), and ~~chat~~ (team commitments, blockers).1112## Sprint / Iteration Planning1314### Goals of Planning15- **Commit to a realistic scope**: What will the team actually ship in this period?16- **Break down work**: Large items become smaller, shippable tickets17- **Surface dependencies**: What blocks what? Who depends on whom?18- **Check capacity**: Does scope fit available time? Who is on vacation, on-call, or ramping?1920### Planning Flow211. **Start with goals**: What should this sprint/iteration achieve? Tie to roadmap or OKR if possible.222. **Review backlog**: Pull from ~~project tracker~~ — what is ready? What is blocked? What has no owner?233. **Break down large items**: Use specs from ~~knowledge base~~ or user input; turn epics/features into tickets.244. **Prioritize**: Order by dependency, value, and urgency. Cut scope if capacity is exceeded.255. **Assign and commit**: Each ticket has an owner and clear acceptance criteria.2627## Breaking Down Work into Tickets2829### When to Create vs. Link30- **Create new tickets** when: Work is not yet in ~~project tracker~~; you are breaking an epic into tasks; you are adding follow-ups or bugs.31- **Link existing tickets** when: Work is already in ~~project tracker~~; you are referencing a parent epic or dependency; you are grouping under a milestone.3233### Good Ticket Titles34- **Specific**: "Add SSO login for Google Workspace" not "SSO support"35- **Action-oriented**: "Implement export to CSV" not "Export feature"36- **One outcome**: One ticket = one shippable result. If it takes multiple PRs, it can still be one ticket; if it is multiple distinct outcomes, split.3738### Good Descriptions39- **Context**: Why is this being done? Link to spec, user story, or customer request.40- **Scope**: What is in scope and what is out of scope for this ticket.41- **Acceptance criteria**: Bullet list of testable conditions that must be true for "done."4243### Acceptance Criteria44- **Testable**: "User can log in with Google Workspace SSO" not "SSO works"45- **Complete**: Cover happy path, main error cases, and edge cases that matter46- **Independent**: Each criterion can be verified on its own47- **Avoid vague words**: "Fast," "user-friendly," "intuitive" — define what they mean concretely4849## Backlog Refinement5051### What to Refine52- **Clarify**: Tickets that are too vague to estimate or implement53- **Split**: Tickets that are too large for one sprint (break into smaller pieces)54- **Merge**: Duplicates or tickets that belong together55- **Prioritize**: Order backlog by value, dependency, and urgency56- **Deprioritize or close**: Items that are no longer relevant5758### Refinement Cadence59- Many teams refine once per sprint (e.g. mid-sprint) so the next sprint planning has a ready backlog.60- Pull from ~~project tracker~~ to see what is in the backlog; use ~~knowledge base~~ for specs and ~~chat~~ for team input on priority.6162## Capacity and Dependencies6364### Capacity Check65- **Who is available?** Subtract PTO, on-call, interviews, other commitments.66- **How much time?** Engineers often have 60–70% of time for planned feature work; the rest is meetings, support, tech debt.67- **Scope vs. capacity**: If backlog exceeds capacity, cut scope. Do not assume people will work overtime.6869### Managing Dependencies70- List dependencies explicitly: "Blocked by ticket X" or "Depends on team Y delivering Z."71- Order work so dependencies are unblocked first.72- Flag cross-team dependencies early; they are the highest risk to schedule.7374## Inputs from Tools7576When planning work or breaking down tickets:7778- **~~project tracker~~**: Backlog items, current capacity, assignees, dependencies, items at risk or blocked79- **~~knowledge base~~**: Specs, PRDs, acceptance criteria, user stories, planning templates80- **~~chat~~**: Team commitments, blockers, capacity notes, recent decisions that affect scope or priority8182If a tool is not connected, use only available data (user-provided backlog, scope, capacity). Note when ~~project tracker~~, ~~knowledge base~~, or ~~chat~~ would improve the plan.