rampstackco/claude-skills / skills/backlog-triage
backlog-triage
VerifiedTurn an inbox of raw requests (bug reports, feature asks, support escalations, feedback, incident follow-ups) into a ranked, deduplicated backlog where every item at the top meets the definition of ready, with recorded scores, merge links, and a written reason for every park and decline. Use this skill whenever the user wants to triage incoming work, groom or prune a backlog, merge duplicate tickets, score or rank requests, decide what the team works on next, or run a weekly triage session. Triggers on triage, groom the backlog, backlog grooming, prioritize the backlog, dedupe tickets, duplicate tickets, what should we work on next, inbox of bugs, feature requests pile, rank these, is this ready, backlog is too big, stale tickets, severity vs frequency. Also triggers when the user pastes a list of raw requests or bug reports and asks what to do with them, even if they never say the word triage.
| File | Kind | Size |
|---|---|---|
| README.md | markdown | 355 B |
| SKILL.md | skill md | 11.6 KB |
| references/aging-policy-guide.md | markdown | 3.3 KB |
| references/example-triage-session.md | markdown | 7.0 KB |
| references/intake-template.md | markdown | 3.2 KB |
| references/scoring-template.md | markdown | 3.4 KB |
| references/triage-checklist.md | markdown | 3.4 KB |
references/intake-template.md
View on GitHub# Intake Template
An intake record is the minimum a request needs before it enters the backlog. It is not a ticket. It is the raw material triage turns into a ticket, a merge, a park, or a decline.
Copy the template, fill every field, and write "unknown" where a field cannot be filled yet. "Unknown" is a question to send, not a blank to ignore.
---
## Template
```
INTAKE ID: [tracker ID or running number]
DATE RECEIVED: [yyyy-mm-dd]
SOURCE: [support queue / sales call / in-app feedback / monitoring alert / incident review / internal request / feedback program]
REPORTER: [name or account, and how to reach them]
REPORTER COUNT: 1
ACCOUNT CONTEXT: [plan, segment, or internal team, if known]
RAW TEXT
[Verbatim. Quote the reporter exactly. Attach screenshots, logs, or call notes.]
AFFECTED AREA
[product area, screen, component, or "unknown"]
FIRST READ (triage lead)
- Likely type: [bug / task / story / spike / epic / unclear]
- Possible duplicates: [IDs, or none found]
- Missing to classify or score: [list, or none]
STATUS: [new / question sent / merged into X / scored / promoted / parked / declined]
```
---
## Field rules
- **Raw text is verbatim.** "User says export is broken" loses the detail. "Clicked Export CSV on the Invoices tab, spinner for 5 minutes, nothing downloaded" is dedupable and reproducible.
- **Reporter must be reachable.** An email, an account ID, or a teammate who owns the relationship. If nobody can be asked a question, the item cannot be made ready.
- **Reporter count starts at 1.** It only goes up through merges. Never estimate it.
- **Affected area can be a guess,** marked as a guess. It drives dedupe search, so a rough area beats none.
---
## Where intake comes from
| Source | Typical shape | What to add at intake |
|---|---|---|
| Support queue | Customer words, often emotional | Account context, ticket link, reproduction attempt by support |
| Sales or success call | Secondhand summary | The customer's own words if recorded, deal or renewal context |
| In-app feedback | Short, anonymous-ish | Account and page from the widget metadata |
| Monitoring alert | Precise, no user context | Affected user count from logs, first-seen date |
| Incident review | Action item | Link to the review, owner, due date from the review |
| Internal request | Often a solution, not a problem | The problem behind the ask. Ask "what happens if we do not do this?" |
| Feedback program | Already themed | The theme's reporter count and source mix from the synthesis |
---
## The question-back message
When an item cannot be classified or scored, send one message with every question at once. Do not drip them.
```
Hi [reporter],
Thanks for flagging [one-line summary]. Before we can prioritize it we need:
1. [Specific question, e.g., "Which browser and which account were you using?"]
2. [Specific question, e.g., "Does this happen every time, or only sometimes?"]
3. [Specific question, e.g., "What were you trying to get done when this happened?"]
We review new items every [day]. If we hear back by then, it goes into that session.
```
Set the item status to "question sent" with the date. If no answer arrives in two sessions, park it with the reason "no response from reporter."
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |