Scrum Conductor
Overview
Facilitates AI-enhanced Scrum orchestration with automated ticket management and high-velocity sprint coordination. Synthesizes daily updates from git activity, detects blockers proactively, and maintains backlog integrity across issue trackers.
When to use: Sprint planning, daily standups, backlog grooming, ticket creation, velocity forecasting, sprint retrospectives, estimation, risk management, release planning, cross-tracker synchronization.
When NOT to use: Technical implementation tasks (use specialized coding skills), architecture design, security auditing.
Quick Reference
| Pattern |
Approach |
Key Points |
| Fact-first standups |
Auto-generate from git logs and PRs |
Never ask humans for data in commit history |
| Ticket engineering |
Machine-readable DoD with acceptance criteria |
Binary true/false criteria, implementation pointers |
| Sprint planning |
Capacity calculation with focus factor and velocity |
Budget = avg velocity (last 3 sprints) adjusted for absences |
| Estimation |
Planning Poker with Fibonacci scale, T-shirt sizing for roadmap |
Reference stories anchor the scale, re-calibrate quarterly |
| Capacity forecasting |
Historical cycle time and lead time |
Factor holidays, context debt, and bottlenecks |
| Backlog grooming |
AI clustering and deduplication |
Flag tickets older than 2 sprints for archive |
| Backlog refinement |
1-2 sessions per sprint, Definition of Ready checklist |
Stories must meet DoR before entering sprint planning |
| Blocker detection |
Scan PRs, assignments, and dependencies |
Flag stale PRs, OOO assignees, breaking deps |
| Parking lot |
Move deep-dives out of standups |
Standups focus on status and blockers only |
| Sprint retrospective |
Start/Stop/Continue with action item tracking |
Track retro action completion rate across sprints |
| Risk management |
Dependency mapping, risk register (probability x impact) |
Address risk scores 6+ immediately, monitor 3-5 |
| Release planning |
Multi-sprint roadmap with confidence levels |
Feature flags decouple deployment from release |
| Escalation tiers |
Tier 1 autonomous, Tier 2 clarification, Tier 3 pairing |
Match response to complexity |
| Priority frameworks |
MoSCoW, value vs effort matrix, WSJF |
WSJF favors small high-value items |
Conductor Protocol
- Ceremony Initialization: Identify the current sprint phase (Planning, Daily, Review, Retro)
- Telemetry Sync: Pull recent activity from git commits, PRs, and communication channels
- Fact Synthesis: Generate factual summaries before ceremonies begin
- Verification: Confirm all action items are converted into tracked tickets with clear owners and DoD
Common Mistakes
| Mistake |
Correct Pattern |
| Asking developers for status updates available in commit logs |
Auto-generate fact summaries from PRs and merges before standups |
| Creating tickets without a machine-readable Definition of Done |
Every ticket needs explicit acceptance criteria and technical pointers |
| Guessing sprint velocity without historical data |
Use cycle time, holidays, and context debt to forecast capacity |
| Letting standups run beyond 15 minutes with deep-dive discussions |
Move deep-dives to a parking lot session; standups focus on blockers |
| Allowing the backlog to grow to 200+ items without pruning |
Auto-flag tickets older than 2 sprints for archive or refactor |
| Using AI to replace human conversation |
Use AI to prepare for the conversation, not substitute it |
| Ignoring team sentiment and morale signals |
High velocity with low morale is a leading indicator of burnout |
| No Definition of Done agreed by the team |
Establish a team-wide DoD checklist applied to every story |
| Skipping retrospectives when the sprint "went fine" |
Every sprint has improvement opportunities; consistency builds the habit |
| Sprint planning without capacity calculation |
Calculate capacity: members x available days x focus factor |
| Stories entering sprint without acceptance criteria |
Enforce Definition of Ready before stories enter a sprint |
| Estimating stories individually instead of as a team |
Use Planning Poker so the full team contributes perspective |
| Slicing stories horizontally by layer |
Slice vertically through all layers so each ticket delivers working functionality |
| No dependency mapping between sprint stories |
Map dependencies explicitly and identify the critical path |
| Retro action items with no owner or due date |
Every action item needs an owner, a due date, and follow-up tracking |
Delegation
- Synthesize daily standup summaries from git activity and PRs: Use
Task agent to pull commit logs and generate fact-based updates
- Cluster and deduplicate backlog tickets across issue trackers: Use
Explore agent to scan GitHub Issues, Jira, and Linear for similar or conflicting items
- Plan sprint capacity and risk assessment using historical velocity: Use
Plan agent to model delivery probability and identify at-risk items
References
- Automated daily rituals, fact-checking workflows, blocker detection, sprint planning, retrospectives, and refinement
- Ticket engineering standards, user story templates, estimation techniques, and priority frameworks
- Predictive velocity, sprint metrics, risk management, distributed teams, and release planning
- Multi-agent task handoffs, escalation tiers, AI retrospectives, blocker detection, and human escalation guardrails
1---2name: scrum-conductor3description: Orchestrates AI-enhanced Scrum ceremonies and sprint coordination. Use when running sprint ceremonies, generating automated daily standups, engineering structured tickets, forecasting sprint capacity, or grooming backlogs. Use for fact-based standups, machine-readable tickets, predictive velocity, and backlog deduplication across GitHub Issues, Jira, and Linear.4license: MIT5---6
7# Scrum Conductor
8
9## Overview
10
11Facilitates AI-enhanced Scrum orchestration with automated ticket management and high-velocity sprint coordination. Synthesizes daily updates from git activity, detects blockers proactively, and maintains backlog integrity across issue trackers.
12
13**When to use:** Sprint planning, daily standups, backlog grooming, ticket creation, velocity forecasting, sprint retrospectives, estimation, risk management, release planning, cross-tracker synchronization.
14
15**When NOT to use:** Technical implementation tasks (use specialized coding skills), architecture design, security auditing.
16
17## Quick Reference
18
19| Pattern | Approach | Key Points |
20| -------------------- | --------------------------------------------------------------- | ------------------------------------------------------------ |
21| Fact-first standups | Auto-generate from git logs and PRs | Never ask humans for data in commit history |
22| Ticket engineering | Machine-readable DoD with acceptance criteria | Binary true/false criteria, implementation pointers |
23| Sprint planning | Capacity calculation with focus factor and velocity | Budget = avg velocity (last 3 sprints) adjusted for absences |
24| Estimation | Planning Poker with Fibonacci scale, T-shirt sizing for roadmap | Reference stories anchor the scale, re-calibrate quarterly |
25| Capacity forecasting | Historical cycle time and lead time | Factor holidays, context debt, and bottlenecks |
26| Backlog grooming | AI clustering and deduplication | Flag tickets older than 2 sprints for archive |
27| Backlog refinement | 1-2 sessions per sprint, Definition of Ready checklist | Stories must meet DoR before entering sprint planning |
28| Blocker detection | Scan PRs, assignments, and dependencies | Flag stale PRs, OOO assignees, breaking deps |
29| Parking lot | Move deep-dives out of standups | Standups focus on status and blockers only |
30| Sprint retrospective | Start/Stop/Continue with action item tracking | Track retro action completion rate across sprints |
31| Risk management | Dependency mapping, risk register (probability x impact) | Address risk scores 6+ immediately, monitor 3-5 |
32| Release planning | Multi-sprint roadmap with confidence levels | Feature flags decouple deployment from release |
33| Escalation tiers | Tier 1 autonomous, Tier 2 clarification, Tier 3 pairing | Match response to complexity |
34| Priority frameworks | MoSCoW, value vs effort matrix, WSJF | WSJF favors small high-value items |
35
36## Conductor Protocol
37
381. **Ceremony Initialization**: Identify the current sprint phase (Planning, Daily, Review, Retro)
392. **Telemetry Sync**: Pull recent activity from git commits, PRs, and communication channels
403. **Fact Synthesis**: Generate factual summaries before ceremonies begin
414. **Verification**: Confirm all action items are converted into tracked tickets with clear owners and DoD
42
43## Common Mistakes
44
45| Mistake | Correct Pattern |
46| ----------------------------------------------------------------- | --------------------------------------------------------------------------------- |
47| Asking developers for status updates available in commit logs | Auto-generate fact summaries from PRs and merges before standups |
48| Creating tickets without a machine-readable Definition of Done | Every ticket needs explicit acceptance criteria and technical pointers |
49| Guessing sprint velocity without historical data | Use cycle time, holidays, and context debt to forecast capacity |
50| Letting standups run beyond 15 minutes with deep-dive discussions | Move deep-dives to a parking lot session; standups focus on blockers |
51| Allowing the backlog to grow to 200+ items without pruning | Auto-flag tickets older than 2 sprints for archive or refactor |
52| Using AI to replace human conversation | Use AI to prepare for the conversation, not substitute it |
53| Ignoring team sentiment and morale signals | High velocity with low morale is a leading indicator of burnout |
54| No Definition of Done agreed by the team | Establish a team-wide DoD checklist applied to every story |
55| Skipping retrospectives when the sprint "went fine" | Every sprint has improvement opportunities; consistency builds the habit |
56| Sprint planning without capacity calculation | Calculate capacity: members x available days x focus factor |
57| Stories entering sprint without acceptance criteria | Enforce Definition of Ready before stories enter a sprint |
58| Estimating stories individually instead of as a team | Use Planning Poker so the full team contributes perspective |
59| Slicing stories horizontally by layer | Slice vertically through all layers so each ticket delivers working functionality |
60| No dependency mapping between sprint stories | Map dependencies explicitly and identify the critical path |
61| Retro action items with no owner or due date | Every action item needs an owner, a due date, and follow-up tracking |
62
63## Delegation
64
65- **Synthesize daily standup summaries from git activity and PRs**: Use `Task` agent to pull commit logs and generate fact-based updates
66- **Cluster and deduplicate backlog tickets across issue trackers**: Use `Explore` agent to scan GitHub Issues, Jira, and Linear for similar or conflicting items
67- **Plan sprint capacity and risk assessment using historical velocity**: Use `Plan` agent to model delivery probability and identify at-risk items
68
69## References
70
71- [Automated daily rituals, fact-checking workflows, blocker detection, sprint planning, retrospectives, and refinement](references/daily-rituals.md)
72- [Ticket engineering standards, user story templates, estimation techniques, and priority frameworks](references/ticket-engineering.md)
73- [Predictive velocity, sprint metrics, risk management, distributed teams, and release planning](references/velocity-risk.md)
74- [Multi-agent task handoffs, escalation tiers, AI retrospectives, blocker detection, and human escalation guardrails](references/agile-agents.md)
oakoss