OPS ► LINEAR COMMAND CENTER

Runtime Context

Before executing, load available context:

1. Secrets: Linear API key required for MCP fallback queries.

   Secret Resolution

   • Check

     $LINEAR_API_KEY

     env var
   • Then: Doppler MCP tools (

     mcp__doppler__*

     ) — if Doppler MCP server is configured
   • Then:

     doppler secrets get LINEAR_API_KEY --plain

     (if doppler CLI configured in prefs)
   • Then: use

     password_manager_config.query_cmd

     from

     ${CLAUDE_PLUGIN_DATA_DIR:-$HOME/.claude/plugins/data/ops-ops-marketplace}/preferences.json

   • If unavailable, use Linear MCP tools exclusively (no curl fallback possible)

2. Preferences: Read

   ${CLAUDE_PLUGIN_DATA_DIR}/preferences.json

   for

   secrets_manager

   /

   doppler

   config.

CLI/API Reference

Linear GraphQL (fallback when MCP unavailable)

curl -X POST https://api.linear.app/graphql -H "Authorization: $LINEAR_API_KEY" -H "Content-Type: application/json" -d '{"query":"{ issues(filter: {state: {type: {in: [\"started\",\"unstarted\"]}}}) { nodes { id title state { name } priority assignee { name } } } }"}'

curl -X POST https://api.linear.app/graphql -H "Authorization: $LINEAR_API_KEY" -H "Content-Type: application/json" -d '{"query":"{ cycles(filter: {isActive: {eq: true}}) { nodes { id number startsAt endsAt } } }"}'

Phase 1 — Load data

Run in parallel:

1. mcp__linear__list_teams

   — get all team IDs

2. mcp__linear__list_issues

   — get issues with cycle filter (use GraphQL fallback for cycle queries if needed)

Then fetch issues for the current cycle:

mcp__linear__list_issues

Route by

$ARGUMENTS

Sprint board view

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 LINEAR ► SPRINT [N] — [start] → [end]
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

IN PROGRESS
 [id]  [priority]  [title]  [assignee]  [estimate]

TODO
 [id]  [priority]  [title]  [assignee]  [estimate]

DONE THIS SPRINT
 [id]  [title]  [completed date]

BLOCKED / CANCELLED
 [id]  [title]  [reason]

──────────────────────────────────────────────────────
 Sprint velocity: [done points] / [total points] ([%])
──────────────────────────────────────────────────────

Use batched AskUserQuestion calls (max 4 options each):

AskUserQuestion call 1:

  [Create new issue]
  [Update issue status]
  [Move issue to/from sprint]
  [More...]

AskUserQuestion call 2 (only if "More..."):

  [View backlog]
  [Sync with GSD phases]

Create issue flow

Collect from user (or parse from

$ARGUMENTS

• Title
• Team (list choices if ambiguous)
• Priority (urgent/high/medium/low)
• Cycle (current sprint or backlog)
• Assignee (optional)
• Estimate (optional)

Use

mcp__linear__create_issue

Created [id]: [title]

GSD sync flow

Read all active GSD STATE.md files across projects. For each active phase:

1. Check if a Linear issue exists with matching phase reference.
2. If not, offer to create one.
3. If status differs (GSD says done, Linear says in-progress), offer to sync.

Update Linear issues to match GSD phase completion status.

Use AskUserQuestion after displaying any view to get the next action.

Native tool usage

Tasks — sprint work tracking

When the user starts working on a Linear issue, use

TaskCreate

TaskUpdate

WebFetch — Linear API fallback

When Linear MCP tools hit quota limits or fail, fall back to

WebFetch

WebFetch(url: "https://api.linear.app/graphql", method: "POST", headers: {"Authorization": "$LINEAR_API_KEY"}, body: '{"query":"{ issues(filter: {cycle: {id: {eq: \"<id>\"}}}}) { nodes { id title state { name } } } }"}')