linear-gather-context - SKILL.md Agent Skill

name: linear-gather-context description: Gather comprehensive Linear context — my tasks, sprint progress, blockers, team workload, prioritized status report target: linear_agent

Linear: Gather Context & Status Report

When to Activate

User wants to know what's happening in Linear — their tasks, sprint progress, blockers, team status, or needs a daily standup summary.

Step 1: My Tasks & Priorities

Get the user's assigned issues:

LINEAR_CUSTOM_GET_MY_TASKS → assigned issues with status, priority, due dates

Or use the standard tool with user filter:

LINEAR_GET_CURRENT_USER → my user ID
LINEAR_LIST_LINEAR_ISSUES(assignee_id="me", first=30)

Categorize by urgency:

OVERDUE / URGENT (priority 1):
  - GEN-142: Fix auth bug (due: yesterday!)
  - GEN-98: Deploy hotfix (priority: Urgent)

DUE THIS WEEK (priority 2-3):
  - GEN-201: API redesign (due: Friday)
  - GEN-187: Update docs (due: Thursday)

BACKLOG / LOW PRIORITY:
  - GEN-250: Refactor utils (no due date)

Step 2: Sprint Progress

Get active sprint/cycle status:

LINEAR_CUSTOM_GET_ACTIVE_SPRINT → current cycle with progress

Or manually:

LINEAR_GET_ALL_LINEAR_TEAMS → team_id
LINEAR_LIST_LINEAR_ISSUES(first=50) → filter by cycle

Sprint summary:

Sprint: "Sprint 23" (Jan 20 - Feb 3)
  Progress: 14/22 issues done (64%)
  Todo: 3 | In Progress: 5 | Done: 14
  
  Velocity: On track (avg 70% at this point)
  
  At Risk:
  - GEN-142: Blocked for 3 days (needs review)
  - GEN-201: Large scope, only 30% done

Step 3: Blockers & Dependencies

Identify blocked issues:

LINEAR_CUSTOM_SEARCH_ISSUES(query="blocked") → blocked issues

For specific issue context:

LINEAR_CUSTOM_GET_ISSUE_FULL_CONTEXT(issue_identifier="GEN-142")
→ Comments, history, sub-issues, relations

Using spawn_subagent for Multiple Issues

When you need to get full context for multiple issues in parallel:

spawn_subagent(task="Get full context for issue GEN-142", context="Extract: comments, history, sub-issues, relations")
spawn_subagent(task="Get full context for issue GEN-201", context="Extract: comments, history, sub-issues, relations")
spawn_subagent(task="Get full context for issue GEN-98", context="Extract: comments, history, sub-issues, relations")

This parallelizes fetching details for multiple issues.

Step 4: Team Context

Get workspace-level view:

LINEAR_CUSTOM_GET_WORKSPACE_CONTEXT → teams, projects, members

Team workload:

LINEAR_LIST_LINEAR_ISSUES(first=50) → all team issues

Step 5: Synthesize Report

Standup format:

Linear Status — Feb 23, 2025

Your Tasks:
  Completed yesterday: GEN-190 (Auth migration)
  In progress: GEN-201 (API redesign, 30%)
  Starting today: GEN-215 (Cache layer)

Sprint 23 (64% complete, 5 days left):
  On track: 14/22 done
  At risk: GEN-142 (blocked), GEN-201 (large)

Blockers:
  - GEN-142: Waiting on review from @alex (3 days)
  
Suggest: Focus on GEN-142 blocker first (overdue), then GEN-201.

Context Resolution

When user references issues by name or vague description:

LINEAR_CUSTOM_RESOLVE_CONTEXT(text="that auth bug") → resolves to GEN-142
LINEAR_CUSTOM_SEARCH_ISSUES(query="authentication error") → search results

Anti-Patterns

Listing all issues without prioritization
Not checking sprint context (user needs timeline awareness)
Ignoring blockers and dependencies
Raw data dump without actionable synthesis
Not resolving vague references to specific issues