next_skill

This skill suggests the most logical next issue to work on by combining cycle planning, dependency analysis, and triage signals.
  • Shell

0

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

3 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill alienfast/claude --skill next

  • SKILL.md3.7 KB

Overview

This skill suggests the best next issue to work on by combining cycle planning, dependency analysis, and triage signals. It prioritizes work that is in the current cycle, newly unblocked, or high priority, and falls back to ready cycle items or backlog quick wins. Use it whenever you ask “what's next”, “next issue”, or run /next.

How this skill works

The skill gathers context from the current cycle and the team dependency graph, and checks for a just-completed issue to identify newly unblocked work. It builds a candidate set of workable issues (not Done/Cancelled/In Progress) whose blockers are resolved and who aren’t actively assigned. Candidates are ranked by a clear tiered order that favors current-cycle and newly-unblocked items, then by priority, cycle membership, remaining blockers, and estimate.

When to use it

  • Starting a fresh day or week and unsure where to begin
  • After finishing an issue (invoked from /finish) to pick the next task
  • Choosing between several candidate issues to maximize momentum
  • When you want to surface transitively unblocked work after clearing a blocker
  • When you need a quick, high-impact issue (quick wins)

Best practices

  • Run with context of a just-completed issue when possible to surface directly and transitively unblocked work
  • Prefer issues in the current cycle that are unblocked before falling back to backlog
  • Respect assignment state: avoid suggesting items someone else is actively working on
  • Use priority > cycle membership > fewer blockers > lower estimate as tiebreakers
  • If the CLI is unauthenticated, authenticate first (linear auth login) to get accurate results

Example use cases

  • You finish PL-14 and invoke /next; the skill recommends PL-17 because it was transitively blocked by PL-14
  • At the start of the sprint you ask 'what's next' and get a cycle-planned, unblocked issue to start work
  • Deciding between two high-priority items: the skill favors the one in the current cycle and with fewer blockers
  • You clear a blocker and the skill finds downstream issues whose last remaining blocker was the one you just closed
  • No clear activity in the cycle: the skill suggests the highest-priority, low-estimate backlog item as a quick win

FAQ

The skill skips cycle-based tiers and ranks workable backlog items by priority and estimate.

What if the linear CLI is not authenticated?

You will be prompted to run linear auth login so the skill can fetch cycle and dependency data.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational