| name | my-issues |
| description | List open issues created by or assigned to the current user, deduplicated and prioritized. |
| context | fork |
My Issues
List all open GitHub issues that are assigned to me or created by me (with assignee empty or myself), deduplicate, and present them sorted by priority.
Arguments
Your args are: $ARGUMENTS
No arguments required.
Steps
Step 1: Determine Current User
ME=$(gh api user --jq '.login')
Step 2: Fetch Issues
Run two queries in parallel:
gh issue list --assignee "$ME" --state open \
--json number,title,labels,assignees,author,createdAt,updatedAt \
--limit 50
gh issue list --author "$ME" --state open \
--json number,title,labels,assignees,author,createdAt,updatedAt \
--limit 50
Step 3: Filter and Deduplicate
- From assigned to me: include all.
- From authored by me: only include if assignee list is empty or contains only me. Exclude issues assigned to other people (they own those issues now).
- Deduplicate by issue number.
Step 4: Classify and Prioritize
Categorize each issue by reading its title, labels, and body (fetch body for each issue). Apply the following priority rules:
| Priority | Criteria |
|---|
| P0 | Bugs or missing features that block core user flows (e.g., broken interactions, missing UI components, data loss) |
| P1 | Important UX issues, newly assigned tasks, features affecting daily user experience |
| P2 | Minor UI polish, experience optimizations, larger features that need scoping |
| P3 | Long-term / operational tasks, nice-to-haves |
Step 5: Output
Use the following format:
## My Open Issues (N total)
### P0 — Blocking Core Functionality
| # | Title | Type | Source | Created |
|---|-------|------|--------|---------|
| #1234 | Short description | Bug | @author → me | 03-19 |
### P1 — Important
...same table format...
### P2 — UX Improvements
...same table format...
### P3 — Long-term / Ops
...same table format...
Column definitions:
- Type: Bug / Feature / Refactor / Ops — inferred from title prefix and labels
- Source: Show as
@author → me if assigned by others, self-created if self-created and self-assigned, self-created (unassigned) if self-created with no assignee
Step 6: Suggested Execution Order
After the table, add a brief Suggested Order section with a numbered list explaining the recommended order and any dependency relationships between issues.
Key Rules
- Use
gh CLI for all GitHub queries
- Always deduplicate by issue number
- Exclude issues created by me but assigned to someone else
- Use English for all output titles and labels
- Fetch issue body (first 500 chars) to make informed priority decisions
- Current user determined via
gh api user --jq '.login'