e2e-test
Write a new Playwright E2E test for Vortex. Accepts a plain-text description of what to test, or a Linear issue ID to fetch the…
Triage open Vortex bug issues in Linear (both community CMTY reports and internal issues). Checks each issue against the codebase and git history, pulls extra context from matching GitHub reports, sizes its reach from ClickHouse telemetry and Mixpanel baselines, and proposes
$ npx -y skills add Nexus-Mods/Vortex --skill triage-issues --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/triage-issuesContext preview
The summary Claude sees to decide when to auto-load this skill.
Triage open Vortex bug issues in Linear (both community CMTY reports and internal issues). Checks each issue against the codebase and git history, pulls extra context from matching GitHub reports, sizes its reach from ClickHouse telemetry and Mixpanel baselines, and proposes
name: triage-issues description: Triage open Vortex bug issues in Linear (both community CMTY reports and internal issues). Checks each issue against the codebase and git history, pulls extra context from matching GitHub reports, sizes its reach from ClickHouse telemetry and Mixpanel baselines, and proposes priorities under the shared Vortex triage policy. Read-only until the user approves. when_to_use: When the user asks to triage Linear bugs or issues for Vortex, re-prioritise a bug backlog, decide what goes into a stable hotfix or the next beta, or check whether a set of issues is still valid, fixed, or duplicated. Not for discovering untracked problems in telemetry — use /triage-telemetry for that. user-invocable: true
Triage existing Vortex bug issues in Linear, using telemetry to size each issue's reach.
Read these shared references before running:
---
Call Linear's `list_issues` with:
Follow `cursor` while `hasNextPage` is true. Keep issues that meet **all** of the following criteria:
Report how many issues were skipped as already triaged.
> [!TIP] > Paginated responses usually exceed the tool output limit and are saved to a local file. Parse that file with a `node -e` one-liner rather than reading the raw file into context.
Resolve `stable`, `beta`, `master`, and new beta features per [Release State in `triage.md`](../_shared/triage.md#release-state).
Follow [Parallel Code Investigation in `triage.md`](../_shared/triage.md#parallel-code-investigation) using the prompt in [`reference.md`](reference.md#agent-prompt). Each subagent also searches GitHub for other users hitting the same bug and pulls in the context they add (versions, games, reproduction steps, logs, workarounds), using only [usable reports](../_shared/github.md#usable-reports). While the background subagents run, proceed to Step 4.
Follow [Sizing Reach and Daily Rate in `triage.md`](../_shared/triage.md#sizing-reach-and-daily-rate):
1. Fetch active-user baselines ([`mixpanel.md`](../_shared/mixpanel.md)) and run the [cause landscape query](../_shared/clickstack.md#cause-landscape-with-8020-cut-used-for-priority-ordering) for stable (and beta) to establish the 80/20 threshold. 2. Map each issue to the specific telemetry causes it covers (matching on the error title, message, or stack trace from the issue description and code check) to score its installs, assign `R1`–`R3`, and calculate its peak daily rate. 3. Flag any widespread causes in telemetry that no existing Linear issue covers—those are candidates for `/triage-telemetry`.
For each issue, combine the code check, GitHub context, and telemetry to determine:
Follow [Presenting and Recording a Triage in `triage.md`](../_shared/triage.md#presenting-and-recording-a-triage), including the close / duplicate / relabel list. Wait for explicit user approval before modifying anything in Linear.
Write a new Playwright E2E test for Vortex. Accepts a plain-text description of what to test, or a Linear issue ID to fetch the…
Triage Vortex error and crash telemetry in ClickStack (ClickHouse) for a release. Ranks error causes against Mixpanel active-user…
Watch or investigate a Vortex log file (rotation- and session-aware). A router over six modes loaded on demand — live tail,…