Source: https://internal-agents.com/agents/linear-agent # Linear — Linear Agent A native agent that synthesizes workspace context, triages, creates follow-up work, runs coding sessions, and executes scheduled/event-driven 'Loops'; used by Linear's own CX, Product, and Engineering teams. - Approach type: Task agent - Deployment stage: Scaled - Autonomy: Drafts reviewed - Evidence strength: Detailed primary - Status: Commercialized - First reported year: 2026 - Work: Support, Customer success, Coding - Interfaces: Slack, Intercom, Linear, Github - Invocation: Event driven, Interactive - Entry reviewed: 2026-09-09 ## Where people stay involved - **assigned coding work → agent-created change** — Work product review · Level 3 ## Overview ### Summary A native agent that synthesizes workspace context, triages, creates follow-up work, runs coding sessions, and executes scheduled/event-driven 'Loops'; used by Linear's own CX, Product, and Engineering teams. Fact · Reported · High confidence · `linear-agent--summary` Confidence reason: A linked first-party source states the claim. Evidence: - Supports · [1] [How we built Linear Agent](https://linear.app/now/how-we-built-linear-agent) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/linear-agent-source-1/content.md) - Supports · [2] [How we use Linear Agent at Linear](https://linear.app/now/how-we-use-linear-agent-at-linear) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/linear-agent-source-2/content.md) - Supports · [6] [Introducing Linear Agent](https://linear.app/changelog/2026-03-24-introducing-linear-agent) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/linear-agent-source-6/content.md) ## How it works ### Supporting component Auto-routes issues, flags duplicates, suggests labels Fact · Reported · High confidence · `linear-agent--primitives-0` Confidence reason: A linked first-party source states the claim. Evidence: - Supports · [2] [How we use Linear Agent at Linear](https://linear.app/now/how-we-use-linear-agent-at-linear) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/linear-agent-source-2/content.md) ### Supporting component Agents get identities, scoped team access, and visible delegation alongside humans Fact · Reported · High confidence · `linear-agent--primitives-1` Confidence reason: A linked first-party source states the claim. Evidence: - Supports · [3] [Our approach to building the Agent Interaction SDK](https://linear.app/now/our-approach-to-building-the-agent-interaction-sdk) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/linear-agent-source-3/content.md) ### Supporting component Scheduled or event-driven agent runs that execute recurring work Fact · Reported · High confidence · `linear-agent--primitives-2` Confidence reason: A linked first-party source states the claim. Evidence: - Supports · [1] [How we built Linear Agent](https://linear.app/now/how-we-built-linear-agent) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/linear-agent-source-1/content.md) ## Supervision evidence ### Operating model assessment Level 3 for assigned coding work → agent-created change; human attention boundary: work-product-review. Inference · Catalog judgment · Medium confidence · `linear-agent--operating-models-0` Confidence reason: The source documents delegated coding output while a human remains accountable for the assigned issue. Qualifications: - Observation date: 2026-08-11 Evidence: - Supports · [1] [How we built Linear Agent](https://linear.app/now/how-we-built-linear-agent) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/linear-agent-source-1/content.md) ## Implementation details ### Sandbox unknown Inference · Catalog judgment · Medium confidence · `linear-agent--architecture-sandbox` Confidence reason: The preserved sources do not document an execution sandbox; unknown does not mean absent. Evidence: - Supports · [1] [How we built Linear Agent](https://linear.app/now/how-we-built-linear-agent) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/linear-agent-source-1/content.md) ### Harness Separate planning agent (triage/issue creation) and coding agent (code generation); Code Intelligence for codebase knowledge; scheduled/event-driven 'Loops' Fact · Reported · High confidence · `linear-agent--architecture-harness` Confidence reason: A linked first-party source states the claim. Evidence: - Supports · [1] [How we built Linear Agent](https://linear.app/now/how-we-built-linear-agent) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/linear-agent-source-1/content.md) - Supports · [2] [How we use Linear Agent at Linear](https://linear.app/now/how-we-use-linear-agent-at-linear) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/linear-agent-source-2/content.md) ### Model Codex was used for internal pull-request review; other model choices are not detailed in the preserved sources Fact · Reported · High confidence · `linear-agent--architecture-model` Confidence reason: A linked first-party source states the claim. Evidence: - Supports · [2] [How we use Linear Agent at Linear](https://linear.app/now/how-we-use-linear-agent-at-linear) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/linear-agent-source-2/content.md) ### Interfaces slack, intercom, linear, github Fact · Reported · High confidence · `linear-agent--architecture-interfaces` Confidence reason: A linked first-party source states the claim. Evidence: - Supports · [2] [How we use Linear Agent at Linear](https://linear.app/now/how-we-use-linear-agent-at-linear) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/linear-agent-source-2/content.md) ### Tool access Triage Intelligence (auto-route, dedup, label); GitHub; testing Code Intelligence + custom MCP servers Fact · Reported · High confidence · `linear-agent--architecture-tool-access` Confidence reason: A linked first-party source states the claim. Evidence: - Supports · [2] [How we use Linear Agent at Linear](https://linear.app/now/how-we-use-linear-agent-at-linear) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/linear-agent-source-2/content.md) ### Knowledge Semantic/vector search evolved into agentic context acquisition across the workspace; Datadog/Sentry customer context Fact · Reported · High confidence · `linear-agent--architecture-knowledge` Confidence reason: A linked first-party source states the claim. Evidence: - Supports · [2] [How we use Linear Agent at Linear](https://linear.app/now/how-we-use-linear-agent-at-linear) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/linear-agent-source-2/content.md) ### Credentials The Agent SDK gives agents explicit identities, scoped OAuth tokens, assignable/mentionable handles, and visible human delegation; issues stay assigned to a human; 'an agent cannot be held accountable' Fact · Reported · High confidence · `linear-agent--architecture-credentials` Confidence reason: A linked first-party source states the claim. Evidence: - Supports · [3] [Our approach to building the Agent Interaction SDK](https://linear.app/now/our-approach-to-building-the-agent-interaction-sdk) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/linear-agent-source-3/content.md) ## Lessons and interpretation ### Lesson Keep the agent close to the source of work (Intercom, Slack, Linear); the best workflows live where work already happens Inference · Catalog judgment · Medium confidence · `linear-agent--lessons-learned-0` Confidence reason: The catalog derives this observation from the linked sources. Evidence: - Supports · [2] [How we use Linear Agent at Linear](https://linear.app/now/how-we-use-linear-agent-at-linear) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/linear-agent-source-2/content.md) ### Lesson Gradual autonomy; start by asking for suggestions, observe, add guidance, only automate once proven reliable Inference · Catalog judgment · Medium confidence · `linear-agent--lessons-learned-1` Confidence reason: The catalog derives this observation from the linked sources. Evidence: - Supports · [2] [How we use Linear Agent at Linear](https://linear.app/now/how-we-use-linear-agent-at-linear) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/linear-agent-source-2/content.md) ### Lesson Break work into small steps to keep coding agents focused and successful Inference · Catalog judgment · Medium confidence · `linear-agent--lessons-learned-2` Confidence reason: The catalog derives this observation from the linked sources. Evidence: - Supports · [2] [How we use Linear Agent at Linear](https://linear.app/now/how-we-use-linear-agent-at-linear) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/linear-agent-source-2/content.md) ### Lesson One Linear engineer reports that agent mistakes reveal possible failure modes during review Inference · Catalog judgment · Medium confidence · `linear-agent--lessons-learned-3` Confidence reason: The catalog derives this observation from the linked sources. Evidence: - Supports · [2] [How we use Linear Agent at Linear](https://linear.app/now/how-we-use-linear-agent-at-linear) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/linear-agent-source-2/content.md) ### Lesson Close the loop; auto-notify the customer when their request ships Inference · Catalog judgment · Medium confidence · `linear-agent--lessons-learned-4` Confidence reason: The catalog derives this observation from the linked sources. Evidence: - Supports · [2] [How we use Linear Agent at Linear](https://linear.app/now/how-we-use-linear-agent-at-linear) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/linear-agent-source-2/content.md) ## Sources 1. [How we built Linear Agent](https://linear.app/now/how-we-built-linear-agent) - Engineering blog · First party · Evidence - Original URL: - Accessed: 2026-08-12 · Last verified: 2026-08-31 - Preserved copy in the repository: 2. [How we use Linear Agent at Linear](https://linear.app/now/how-we-use-linear-agent-at-linear) - Engineering blog · First party · Evidence - Original URL: - Accessed: 2026-08-12 · Last verified: 2026-08-31 - Preserved copy in the repository: 3. [Our approach to building the Agent Interaction SDK](https://linear.app/now/our-approach-to-building-the-agent-interaction-sdk) - Engineering blog · First party · Evidence - Original URL: - Accessed: 2026-08-12 · Last verified: 2026-08-31 - Preserved copy in the repository: 4. [Hacker News submission for how Linear built its agent](https://news.ycombinator.com/item?id=49252304) - Hn thread · Community · Discovery - Original URL: - Published: 2026-08-11 · Accessed: 2026-08-12 · Last verified: 2026-08-31 - Preserved copy in the repository: 5. [Hacker News submission for the Linear Agent public beta](https://news.ycombinator.com/item?id=48503334) - Hn thread · Community · Discovery - Original URL: - Published: 2026-06-12 · Accessed: 2026-08-12 · Last verified: 2026-08-31 - Preserved copy in the repository: 6. [Introducing Linear Agent](https://linear.app/changelog/2026-03-24-introducing-linear-agent) - Release · First party · Evidence - Original URL: - Published: 2026-03-24 · Accessed: 2026-08-12 · Last verified: 2026-08-31 - Preserved copy in the repository: