Automatically maintains project BACKLOG.md with deferred features, tech debt, bugs, and improvements. Adds items when work is deferred, marks complete when addressed. Used by other skills to track out-of-scope work.
Automatically maintains a BACKLOG.md file tracking deferred work, tech debt, and bugs. Activates when work is deferred during development or when you query/add items manually.
/plugin marketplace add astrosteveo/harness-mp/plugin install harness@harness-mpThis skill inherits all available tools. When active, it can use any tool Claude has access to.
backlog-template.mdAutomatically maintain a living BACKLOG.md document tracking deferred work, tech debt, bugs, and improvements.
Core principle: Backlog maintains itself. Users only interact to query status or make manual adjustments.
Automatic (no user prompt):
Manual (user request):
Location: Project root (BACKLOG.md)
Organization:
Format:
# Project Backlog
## High Priority
### Features
- [ ] Feature description (added: 2026-01-05)
### Tech Debt
- [ ] Refactor X (added: 2026-01-04)
### Bugs
- [ ] Fix Y (added: 2026-01-05)
### Improvements
- [ ] Optimize Z (added: 2026-01-03)
## Medium Priority
[same subsections...]
## Low Priority
[same subsections...]
---
## Completed
### Features
- [x] User profile (completed: 2026-01-03, commit: 3df7661)
### Tech Debt
- [x] Refactor auth (completed: 2026-01-04, commit: b82e439)
### Bugs
- [x] Fix race condition (completed: 2026-01-05, commit: c91a023)
### Improvements
- [x] Optimize queries (completed: 2026-01-02, commit: f4e8a12)
1. Check for BACKLOG.md
skills/managing-backlog/backlog-template.mdchore: initialize project backlog2. Check for duplicates
3. Determine priority and type
Use context to set smart defaults:
| Context | Priority | Type |
|---|---|---|
| Bug found during debugging | Medium | Bugs |
| Deferred feature from brainstorming | Low | Features |
| Tech debt from code review (Minor) | Low | Tech Debt |
| Tech debt from code review (Important) | Medium | Tech Debt |
| Performance improvement noted | Low | Improvements |
| Security issue found | High | Bugs |
| Breaking change needed | High | Tech Debt |
4. Insert item
- [ ] {description} (added: YYYY-MM-DD)5. Announce and commit
chore: add backlog item - {description}During code review, you find magic numbers that should be constants.
You: "Adding to backlog as [Low Priority/Tech Debt]: Replace magic numbers with named constants in validation module"
[Updates BACKLOG.md automatically]
[Commits change]
Malformed BACKLOG.md:
Git conflicts:
Explicit reference (preferred):
Fuzzy matching (secondary):
1. Identify matching item
2. Mark complete and move
- [ ] → - [x](completed: YYYY-MM-DD, commit: SHA)3. Get commit SHA
For git projects:
git rev-parse HEAD | cut -c1-7
For non-git projects:
(completed: YYYY-MM-DD)4. Announce and commit
chore: mark backlog item complete - {description}Check for completed backlog items:
You just implemented caching for API responses.
Backlog has: "- [ ] Add caching layer to API responses (added: 2026-01-03)"
You: "Marking backlog item complete: Add caching layer to API responses"
[Moves item to Completed section]
Item becomes: "- [x] Add caching layer to API responses (completed: 2026-01-05, commit: a7f3c21)"
[Commits change]
Users can manually interact with the backlog through these commands:
"Show me the backlog"
"What's high priority?" / "What's medium priority?" / "What's low priority?"
"What bugs are in the backlog?" / "What features..." / "What tech debt..." / "What improvements..."
"What's the backlog status?"
"Add X to backlog as [priority/type]"
"Remove Y from backlog"
chore: remove backlog item - {description}"Mark Z complete"
"Change X to [priority]"
chore: reprioritize backlog item - {description}User: "What's high priority?"
You: "High priority items in backlog:
Features:
- Real-time notifications (added: 2026-01-04)
Bugs:
- Fix session cleanup (added: 2026-01-05)
2 total high priority items."
User: "Add support for PostgreSQL to backlog as High Priority Feature"
You: "Adding to backlog as [High Priority/Feature]: Support for PostgreSQL"
[Updates BACKLOG.md]
[Commits]
Creating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Use this when users request creating art using code, generative art, algorithmic art, flow fields, or particle systems. Create original algorithmic art rather than copying existing artists' work to avoid copyright violations.
Applies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual formatting, or company design standards apply.
Create beautiful visual art in .png and .pdf documents using design philosophy. You should use this skill when the user asks to create a poster, piece of art, design, or other static piece. Create original visual designs, never copying existing artists' work to avoid copyright violations.