AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Connector Linear

skill-sboghossian-mini-claude-for-legal-connector-linear · by sboghossian

Use when the legal-AI product team needs to read, create, or update issues in the Linear project-management system — including automated bug report creation from chat, feature-request capture, sprint status queries, and pull-request linking. An internal engineering and ops connector; not a legal-content tool. Triggers on requests to create a bug or feature request, check sprint status, or link a…

No reviews yet
0 installs
28 views
0.0% view→install

Install

$ agentstack add skill-sboghossian-mini-claude-for-legal-connector-linear

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-sboghossian-mini-claude-for-legal-connector-linear)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
4mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Connector Linear? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Connector — Linear

What it does

Linear is the issue-tracker and project-management tool for the legal-AI product team. This connector is the integration layer between the AI assistant, the product's automated monitoring, and the engineering team's Linear workspace.

Core functions:

  • Read — retrieve issue details, search issues by query, list team sprints and project status.
  • Write — create bug reports and feature requests from chat-captured signals, update issue status.
  • Watch — monitor issue status changes and surface updates to relevant users.

This connector is used internally by the platform team — it is not exposed to end-user lawyers as a legal-workflow tool.

Setup / auth

Two authentication methods are supported:

OAuth 2.0 (preferred for user-scoped actions)

  • User authorizes via Linear OAuth flow.
  • Access token scoped to the user's teams and workspace.
  • Suitable when an engineer or PM needs to perform actions under their own identity.

Personal Access Token (PAT — for service accounts)

  • Generate a PAT in Linear Settings → API.
  • Scope to the specific teams and projects the service account needs.
  • Used for automated actions (bug report creation, status monitoring).
  • Stored in the platform's secrets manager; rotated every 90 days.

Workspace configuration

  1. Select the Linear workspace on initial connector setup.
  2. Map team identifiers to their workflow purposes (e.g., "DES" team = design, "ENG" team = engineering, "OPS" team = operations).
  3. Configure webhook subscriptions for the issue events the platform should monitor.

Capabilities

| Capability | Auth required | Notes | |---|---|---| | Read issue by ID | Read | Returns full issue detail including description, labels, assignee, status | | Search issues | Read | Full-text search + filter by label, assignee, project, status | | List team members | Read | Useful for assignee suggestions | | List teams and projects | Read | For routing new issues to the right team | | Create new issue | Write | With title, description, team, label, priority | | Update issue status | Write | Move through the team's workflow states | | Add comment to issue | Write | For logging automation-generated notes | | Create sub-issue | Write | For breaking down a bug into sub-tasks | | Watch issue for status changes | Webhook | Subscribe to issueUpdated events | | Link pull request | Write via PR integration | Associate a GitHub PR with a Linear issue |

Use cases

Bug report collection (from chat)

When a user reports a product issue in chat (e.g., "the NDA generator crashed mid-generation"), the assistant:

  1. Asks one clarifying question: "What browser / device, and what were you doing just before the crash?"
  2. Creates a Linear issue in the ENG team with:
  • Title: formatted bug title.
  • Description: user-provided context + session ID + timestamp.
  • Label: bug + severity (P0/P1/P2 based on impact).
  • Assignee: on-call engineer (from a configured rotation list).
  1. Returns the Linear issue URL to the user with "I've logged this as ENG-[N]. Our team will follow up within [SLA]."

Feature request collection

When a user expresses a feature wish ("I wish Louis could auto-file documents with the DIFC Court"):

  1. Confirm interest: "Should I log this as a feature request for the product team?"
  2. On confirmation, create a Linear issue in the PROD team with:
  • Title: feature request summary.
  • Description: verbatim user request + context.
  • Label: feature-request + relevant practice area tag.
  1. Return the issue URL.

Do not over-promise. Never say the feature will be built — say it will be reviewed.

Sprint planning integration

Retrieve the current sprint status for a team:

  • List issues in the current cycle with their status (In Progress / In Review / Done / Backlog).
  • Surface any overdue issues (due date passed + not Done).
  • Flag blockers (issues with the blocked label).

Useful for daily standups or asynchronous status digests.

Status dashboard

Aggregate issue counts per label and per status for a product health view:

  • Open bugs by severity.
  • Feature requests by vote count or recency.
  • Deployment blockers.

Pull request linking

When a PR is opened in GitHub that references a Linear issue (via the ENG-123 convention in the PR title or branch name), the connector can:

  • Look up the Linear issue.
  • Update its status to "In Review."
  • Add a comment linking the PR URL.

Permissions model

| User type | Linear access level | Permitted actions | |---|---|---| | Chat users (end-users) | None (create via service account only) | Submit bug/feature reports; view issue URL | | Platform engineers | Team member | Full read/write on their team | | Senior engineers / PMs | Team admin | Cross-team visibility; priority changes | | Admin | Workspace admin | Global; required for webhook setup and integration config |

A user should never be able to read another user's private issue comments or attached files they are not authorized to see.

Tenant isolation

Each tenant connects their own Linear workspace:

  • The connector stores the workspace ID per tenant.
  • All API calls from a given tenant are scoped to their workspace.
  • Cross-tenant issue creation or reading is prohibited and treated as a security incident.

Audit logging

All chat-driven Linear actions are logged:

  • Actor: user ID (for user-initiated) or service account (for automated).
  • Action: create / update / comment.
  • Issue ID and team.
  • Timestamp (UTC).

Audit logs are retained for 90 days.

Failure modes

| Failure | Cause | Resolution | |---|---|---| | Unauthorized | PAT expired or revoked | Rotate PAT; alert platform engineer | | Team not found | Team ID changed (Linear allows renaming) | Refresh team list from workspace; update mapping | | Issue creation fails | Description too long (>64KB) | Truncate description; attach full context as a comment | | Webhook not delivered | Network interruption | Linear retries for 24h; implement a reconciliation job to catch missed events | | Duplicate issue | Same bug reported by multiple users | Implement deduplication by checking for open issues with similar title in the last 7 days before creating |

Related skills

  • [[connector-hubspot-crm]]
  • [[connector-posthog]]
  • [[connector-figma]]

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.