Stats
Actions
Tags
From claude-code-skills
Proactively identify at-risk issues and PRs: stale items, blocked work, deadline risks, scope creep, and unassigned high-priority items
How this skill is triggered — by the user, by Claude, or both
Slash command
/claude-code-skills:at-risk-detectionThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Proactively scan GitHub issues and pull requests to identify at-risk items before they become problems.
Proactively scan GitHub issues and pull requests to identify at-risk items before they become problems.
Activate when:
Fetch the 30 most recent open issues:
Fetch ALL open issues by paginating:
Signal: No activity for extended period Default Threshold: 14 days for issues, 3 days for PRs awaiting review Severity Escalation:
How to detect: Check updated_at field on each issue/PR. Calculate days since last update.
Signals:
blocked, waiting, on-hold, waiting-for-responseHow to detect: Check labels array for blocking keywords. Search body text for blocking phrases.
Signal: Milestone deadline approaching or overdue Severity Levels:
How to detect: Check if issue has a milestone with a due_on date. Compare to current date.
Signals:
How to detect: Check comments count on issues. For PRs, check additions + deletions if available.
Signal: High-priority issue without assignee
Priority Labels: priority:high, urgent, P1, P0, critical, blocker
How to detect: Check if assignee is null AND labels contain any priority keywords.
## Project Health: [Project Name]
### Summary
- **Total Open Items**: X (Y issues, Z PRs)
- **At-Risk Items**: N
- **Health Score**: X%
### Risk Breakdown
- Critical: X
- Warning: Y
- Info: Z
### Overloaded Assignees (if any)
- @username: N assigned issues
## At-Risk: #[number] - [title]
**Severity**: [Critical | Warning | Info]
**Type**: [Issue | PR]
**URL**: [link]
**Assignee**: @username
### Risk Signals
- [signal type]: [description]
### Suggestions
- [actionable recommendation]
For each risk signal detected, provide actionable suggestions:
| Risk Type | Suggestion |
|---|---|
| Stale issue | "Comment with status update or close if no longer relevant" |
| Stale PR | "Request review or close if superseded" |
| Blocked | "Resolve blocking dependency or re-prioritize" |
| Deadline approaching | "Break into smaller milestones or adjust deadline" |
| Scope creep | "Split into smaller issues or create a tracking epic" |
| Unassigned high-priority | "Assign an owner immediately" |
| Overloaded assignee | "Redistribute work or adjust priorities" |
Track issue counts per assignee:
| Setting | Default | Description |
|---|---|---|
| staleDays | 14 | Days before issue is stale |
| prStaleReviewDays | 3 | Days before PR awaiting review is stale |
| deadlineWarningDays | 7 | Days before deadline to warn |
| maxCommentsBeforeScopeWarning | 30 | Comment threshold for scope creep |
| maxAssignedIssuesPerPerson | 10 | Max issues before overload warning |
npx claudepluginhub britt/claude-code-skills --plugin writing-user-storiesTriages GitHub issues assigned to you: fetches via GH CLI with cc-triage-scope label, builds dependency graph, categorizes resolved issues, and applies labels (cc-create-issue, cc-exec-issue, cc-update-issue).
Provides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.