Streamline Linear Workflows with the Linear Agent Skill

Where this fits

You're mid-sprint, and your agent needs to close a completed issue, pull open tasks for a project, or move a bug into the next cycle. Instead of you switching to Linear and manually syncing state, this skill gives the agent a direct, scriptable interface to your workspace. It's designed for developers who want AI agents to handle Linear reads and writes without exposing API keys or building custom GraphQL wrappers.

Why agents benefit

• Reads are cached — list and get operations hit a local cache first, so agents avoid repeated round-trips and stay fast during multi-step tasks.
• Writes are live and consistent — create, update, complete, and archive operations write through to the cache, so a follow-up read reflects the change immediately.
• Broad resource coverage — issues, projects, cycles, milestones, initiatives, and labels are all accessible from one CLI, reducing the need for multiple integrations.
• Per-resource help — linear.sh <resource> --help shows available actions, letting the agent discover syntax without guessing or reading the full README.
• Self-service error reporting — problems file automatically via kendex report, so the agent can flag broken behavior without manual triage.

Practical scenarios

Sprint cleanup without context switching

An agent lists all issues in the current cycle, checks which are marked complete, and archives the ones with no activity. It then updates the cycle description with a summary—all without you opening the Linear app.

Cross-project dependency checks

When a new issue is created, the agent looks up related projects and labels, then adds a relation to a parent issue. It validates completion criteria first, so the new task is correctly scoped from the start.

Bulk updates for triage

Your agent fetches all unassigned issues in a project, assigns them based on labels, and creates a comment explaining each assignment. A quick bulk-update handles the changes in one call, then the cache refreshes for the next query.

Add it to your agent workflow

Install the skill with:

npx skills add vanillagreencom/kendex --skill linear

Then call it from your agent runtime:

.agents/skills/linear/scripts/linear.sh issues list --project "Mobile App" --state "In Progress"

This lists all in-progress issues for the Mobile App project, using the cache for a quick response.

Compare before adopting

Look at the skill's maintenance freshness—version 1.1.0 and an active repo suggest ongoing updates. Check the quality signals: stars (65), a real MIT license, and documented error reporting. Compare it with custom Linear GraphQL scripts or other agent skills that may cover only issues or only reads. Make sure the skill fits your workflow: does your agent already manage Linear through another tool? Would the local cache conflict with your existing data source?

Why it is worth tracking

This skill passes a practical bar: it covers the core Linear resources, keeps read performance viable with caching, and is actively maintained. The modest star count is less important than the clean CLI and the built-in report path, which signal a developer who cares about real use. It's worth evaluating now if Linear is a regular part of your agent workflows—see more details on the skill page to track its evolution and community feedback.