Agent skill · Audited from source

cyrus-setup-linear

skill · ceedaragents/cyrusApache-2.0 licenceAudited 26 Oct 2026 · source a090b6160be5

★ 872pushed today
✕BLOCKGrade D · 69/100 · 6 findings · a090b61
872repo stars · whole ceedaragents/cyrus

Is it safe? No — BLOCK, grade D (69/100). We read this skill's source at commit a090b6160be5 and found no filesystem, network or shell access and no hard-coded secrets. The audit raised 6 findings, 1 critical or high. What it does: Create a Linear OAuth application and configure Cyrus to use it — supports agent-browser automation or guided manual setup.

Ask an AI about this pageGChatGPTCClaudePPerplexityGeGeminiXGrok
More
Cite

Cite this page

ModelsAtlas. “cyrus-setup-linear (ceedaragents/cyrus)”. Retrieved 26 Oct 2026. https://modelsatlas.com/skills/ceedaragents/cyrus-setup-linear

Blocked. The page stays public so you can check this skill if someone recommended it to you. The audit found 1 critical or high finding, starting with “Prompt injection” in SKILL.md:172. Do not install it without reading them.

2 skills are called “cyrus-setup-linear”. You are viewing ceedaragents/cyrus. 2 of them are BLOCK.

Trust audit

Audit 0.4.1 · source read 26 Oct 2026

Trust score69/100✕BLOCK D
0 critical1 high5 medium0 low0 info

BLOCK means the audit found something that could harm you or your machine. Grade D is shared by 642 of 14,816 audited skills.

What the source does

  • FilesystemNone observedNo reads or writes in the source
  • NetworkNone observedNo outbound calls in the source
  • ShellNone observedNo commands run in the source
  • DependenciesPinnedNothing floats
  • SecretsNone foundNo keys in source

Five audit layers

LayerWhat we checkResultFindings
L0 ProvenanceRepo owner, licence, archive status, activityWARN5
L1 Static sourceSecrets, path traversal, supply chain, hidden codeNA0
L2 InstructionsPrompt injection, authority framing, hidden directivesFAIL1
L3 InventoryThe files and declarations match the sourcePASS0
L4 SandboxRun in isolation and watch behaviourSKIPPED why0

Gates applied: no behavioural pass.

Findings (6)

HIGHPrompt injectionprompt.credential_read · SKILL.md:172
F-01 Prompt injectionCWE-94 · CWE-1427Layer L2
## Step 4: Collect Credentials
Why
asks the agent to read credentials
Fix
—

Audit 0.4.1 · source sha a090b6160be5 · 26 Oct 2026Report an issue or request a re-scan

Provenance

Who published it, and what the verdict is pinned to

  • Publisherceedaragents GitHub organizationOwner of the repository this skill is read from
  • Source folderceedaragents/cyrus/skills/cyrus-setup-linearThe exact folder we audited
  • LicenceApache-2.0As the repository states it
  • Commit pinned for this verdicta090b6160be5A new audit runs when the source changes
  • Repository statusNot archivedPushed today
  • Sandboxed runNot runSkills are read, not executed; layer L4 is skipped

What it tells the agent

Full source on GitHub ↗

Its description: Create a Linear OAuth application and configure Cyrus to use it — supports agent-browser automation or guided manual setup.

---
name: cyrus-setup-linear
description: Create a Linear OAuth application and configure Cyrus to use it — supports agent-browser automation or guided manual setup.
---

**CRITICAL: Never use `Read`, `Edit`, or `Write` tools on `~/.cyrus/.env` or any file inside `~/.cyrus/`. Use only `Bash` commands (`grep`, `printf >>`, etc.) to interact with env files — secrets must never be read into the conversation context. Never scrape, extract, or read secret values from web pages — guide the user to copy them manually.**

# Setup Linear

Creates a Linear OAuth application and configures credentials so Cyrus can receive webhooks and respond to issues.

## Step 1: Check Existing Configuration

```bash
grep -E '^LINEAR_CLIENT_ID=' ~/.cyrus/.env 2>/dev/null
```

If `LINEAR_CLIENT_ID` is already set, check if OAuth is also complete:

```bash
grep -q '"workspaces"' ~/.cyrus/config.json 2>/dev/null && echo "configured" || echo "not configured"
```

If both are set, inform the user:

> Linear is already configured. Skipping this step.
> To reconfigure, remove `LINEAR_CLIENT_ID`, `LINEAR_CLIENT_SECRET`, and `LINEAR_WEBHOOK_SECRET` from `~/.cyrus/.env` and re-run.

Skip to completion.

## Step 2: Get CYRUS_BASE_URL

Read the base URL from the env file (set by `setup-endpoint`):

```bash
grep '^CYRUS_BASE_URL=' ~/.cyrus/.env | cut -d= -f2-
```

This is needed for the callback and webhook URLs.

## Step 3: Create Linear OAuth App From Manifest

Linear OAuth apps must be created from a manifest-backed setup URL. Do not create the app from scratch or fill each field manually. The manifest flow keeps the app configuration reproducible and avoids drift in callback, webhook, and event type settings.

### 3a. Build the manifest URL

Generate a JSON OAuth app manifest and encode it into Linear's `manifest` query parameter:

```bash
CYRUS_BASE_URL="<CYRUS_BASE_URL>" \
AGENT_NAME="<AGENT_NAME>" \
AGENT_DESCRIPTION="<AGENT_DESCRIPTION>" \
node <<'NODE'
const fs = require("node:fs");

const baseUrl = process.env.CYRUS_BASE_URL.replace(/\/+$/, "");
const agentName = process.env.AGENT_NAME || "Cyrus";
const agentDescription =
	process.env.AGENT_DESCRIPTION || "AI coding agent for automated development";

const manifest = {
	$schema: "https://linear.app/.well-known/oauth-app-manifest.schema.json",
	schemaVersion: "1.0.0",
	distribution: "private",
	display: {
		description: agentDescription,
	},
	developer: {
		name: "Self-hosted",
	},
	oauth: {
		client_name: agentName,
		client_uri: "https://github.com/ceedaragents/cyrus",
		redirect_uris: [`${baseUrl}/callback`],
		grant_types: ["authorization_code"],
	},
	webhook: {
		enabled: true,
		url: `${baseUrl}/linear-webhook`,
		resourceTypes: [
			"AgentSessionEvent",
			"AppUserNotification",
			"PermissionChange",
			"Issue",
		],
	},
};

const manifestJson = JSON.stringify(manifest, null, 2);
const manifestUrl = `https://linear.app/settings/api/applications/new?manifest=${encodeURIComponent(JSON.stringify(manifest))}`;

fs.writeFileSync("/tmp/cyrus-linear-oauth-app-manifest.json", `${manifestJson}\n`);
fs.writeFileSync("/tmp/cyrus-linear-oauth-app-url.txt", `${manifestUrl}\n`);

console.log(manifestJson);
console.log(`\n${manifestUrl}`);
NODE

LINEAR_MANIFEST_URL="$(cat /tmp/cyrus-linear-oauth-app-url.txt)"
```

The same URL is now stored in `LINEAR_MANIFEST_URL` for automation commands and in `/tmp/cyrus-linear-oauth-app-url.txt` for manual copy/paste. The JSON is stored in `/tmp/cyrus-linear-oauth-app-manifest.json` for review/debugging only; do not ask the user to fill the OAuth app form manually from it.

If `<AGENT_NAME>` contains the word "Linear" or a URL, choose a different OAuth application name before building the manifest. Linear rejects OAuth client names with those values.

Determine which browser automation mode to use (see orchestrator rules):

1. If `claude-in-chrome` MCP tools are available → use **Path A-1** (claude-in-chrome)
2. If `agent-browser` is installed (`which agent-browser`) and a Chrome debug session is connected → use **Path A-2** (agent-browser)
3. Otherwise → use **Path B** (manual)

### Path A-1: claude-in-chrome Automation

Use the `mcp__claude-in-chrome__*` tools to navigate and interact with the user's existing Chrome browser. The user is likely already signed in to Linear.

Navigate to `LINEAR_MANIFEST_URL`. The page should open with the manifest-applied OAuth app configuration already populated.

Review only non-secret fields:

- Application name matches `<AGENT_NAME>`
- Developer name is `Self-hosted`
- Developer URL is `https://github.com/ceedaragents/cyrus`
- Redirect callback URLs contains `<CYRUS_BASE_URL>/callback`
- Webhook is enabled with URL `<CYRUS_BASE_URL>/linear-webhook`
- Event types include Agent session events, Inbox notifications, Permission changes, and Issues
- Public/distribution is private or disabled

If the user is not signed in, pause and ask them to sign in. After review, click **Create**. **Do NOT screenshot credential pages or attempt to scrape secrets.** Proceed to Step 4.

### Path A-2: agent-browser Automation

If `agent-browser` is connected to a Chrome debug session, automate the Linear app creation.

#### 3b. Navigate to the manifest URL

```bash
agent-browser navigate "$LINEAR_MANIFEST_URL"
```

Wait for page to load. Take a screenshot to verify you're on the right page and logged in.

#### 3c. Review and create

Review the same non-secret manifest-applied fields listed in Path A-1. If the page asks the user to sign in, pause and let them complete sign-in. Click **Create**.

After creation, Linear redirects to the app settings page. **Do NOT screenshot credential pages or attempt to scrape secrets.** Proceed to Step 4.

### Path B: Manual Guided Setup

Guide the user through manual creation:

> ### Create a Linear OAuth Application
>
> 1. Open this manifest-backed Linear app creation URL:
>
>    `<LINEAR_MANIFEST_URL>`
>
> 2. Sign in to Linear if prompted, then review the pre-filled settings:
>    - *

Excerpt, first 6,000 characters of SKILL.md, verbatim from the audited commit. It is text from someone else’s repository, quoted so you can judge it before your agent reads it.

Install

Pinned to the audited commit a090b61

git clone https://github.com/ceedaragents/cyrus
git -C cyrus checkout a090b6160be5

The author documents no install command, so this uses the repository directly. Our verdict covers commit a090b6160be5 only.

Works in

The author has not said which agents this skill supports, and we have not tested it in any.

Audit history

Every audit, every source change

  1. ✕BLOCK D 69 / 100

    first audit · source a090b6160be5

Similar skills

Questions about cyrus-setup-linear

What does the cyrus-setup-linear skill do?
The Claude Code background agent for Linear, Slack, Github, GitLab etc. you deploy anywhere. Supports Codex, Cursor, Gemini, and Opencode harnesses too.
Is cyrus-setup-linear safe to install?
No — not without reading the findings first. The audit graded it D (69/100) and found 1 critical or high issue in the source. Each one is listed on this page with the file and line it is on.
What can cyrus-setup-linear access on my machine?
The audit observed no filesystem, network or shell use at all in its source.
How current is this page?
The grade is for one exact copy of the source (a090b6160be5), read on 2026-10-26. The repository is watched, and a new audit runs when it changes — this is the first audit.

Sources

Spotted something wrong? Report a correction · Methodology

Keep going in your AI

Opens with this page as the source. We add nothing else to the prompt.