Agent skill · Audited from source

cyrus-setup-claude-auth

skill · ceedaragents/cyrusApache-2.0 licenceAudited 21 Oct 2026 · source 95804c41da68

★ 859pushed today
✕BLOCKGrade D · 69/100 · 6 findings · 95804c4
859repo stars · whole ceedaragents/cyrus

Is it safe? No — BLOCK, grade D (69/100). We read this skill's source at commit 95804c41da68 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: Configure Claude Code authentication for Cyrus — API key, OAuth token, or third-party provider.

Ask an AI about this pageGChatGPTCClaudePPerplexityGeGeminiXGrok
More
Cite

Cite this page

ModelsAtlas. “cyrus-setup-claude-auth (ceedaragents/cyrus)”. Retrieved 21 Oct 2026. https://modelsatlas.com/skills/ceedaragents/cyrus-setup-claude-auth

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:98. Do not install it without reading them.

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

Trust audit

Audit 0.4.1 · source read 21 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:98
F-01 Prompt injectionCWE-94 · CWE-1427Layer L2
read -s -p "Paste your Anthropic API key: " val && printf 'ANTHROPIC_API_KEY=%s\n' "$val" >> ~/.cyrus/.env && echo " ✓ Saved"
Why
asks the agent to read credentials
Fix
—

Audit 0.4.1 · source sha 95804c41da68 · 21 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-claude-authThe exact folder we audited
  • LicenceApache-2.0As the repository states it
  • Commit pinned for this verdict95804c41da68A 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: Configure Claude Code authentication for Cyrus — API key, OAuth token, or third-party provider.

---
name: cyrus-setup-claude-auth
description: Configure Claude Code authentication for Cyrus — API key, OAuth token, or third-party provider.
---

**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.**

# Setup Claude Auth

Configures Claude Code credentials so Cyrus can run AI sessions.

## Step 1: Check Existing Auth

Check if credentials are already configured:

```bash
grep -c -E '^(ANTHROPIC_API_KEY|CLAUDE_CODE_OAUTH_TOKEN)=' ~/.cyrus/.env 2>/dev/null || echo "0"
```

If the count is >= 1, inform the user:

> Claude Code authentication is already configured. Skipping this step.
> To reconfigure, remove the existing key from `~/.cyrus/.env` and re-run this skill.

Skip to completion.

## Step 2: Choose Auth Method

Ask the user:

> **How would you like to authenticate Claude Code?**
>
> 1. **Current account** (easiest) — use the credentials from your active `claude` CLI session
> 2. **API Key** — from [console.anthropic.com](https://console.anthropic.com/)
> 3. **Separate OAuth token** — run `claude setup-token` to generate a token for a specific account
> 4. **Third-Party Provider** — Vertex AI, AWS Bedrock, Azure, etc.

## Step 3: Configure Credentials

**CRITICAL: Secrets must NEVER appear in the conversation.** Do not explore `~/.claude/` looking for credential files. Use only the methods below.

Detect the OS for the right clipboard command:

```bash
uname -s
```

### Option 1: Current Account

Check if `claude` is authenticated on this machine:

```bash
claude auth status
```

If authenticated, instruct the user to run `claude setup-token` which will output a token.

To capture it safely, **append a placeholder line** to the env file, then open it in a GUI window so the user can paste the token:

```bash
grep -q '^CLAUDE_CODE_OAUTH_TOKEN=' ~/.cyrus/.env || echo 'CLAUDE_CODE_OAUTH_TOKEN=' >> ~/.cyrus/.env
```

Then open the file in a separate window — use whatever GUI editor is available:

```bash
# macOS: VS Code if available, otherwise TextEdit
code --new-window ~/.cyrus/.env 2>/dev/null || open -a TextEdit ~/.cyrus/.env

# Linux
code ~/.cyrus/.env 2>/dev/null || xdg-open ~/.cyrus/.env
```

Tell the user:

> 1. Run `claude setup-token` in a separate terminal
> 2. Copy the token it outputs
> 3. I've opened `~/.cyrus/.env` — find the `CLAUDE_CODE_OAUTH_TOKEN=` line
> 4. Paste the token right after the `=` (no spaces, no newline)
> 5. Save and close the file

### Option 2: API Key

Instruct the user to copy their API key from the [Anthropic Console](https://console.anthropic.com/), then provide the appropriate command:

**macOS:**
```bash
printf 'ANTHROPIC_API_KEY=%s\n' "$(pbpaste)" >> ~/.cyrus/.env
```

**Linux:**
```bash
printf 'ANTHROPIC_API_KEY=%s\n' "$(xclip -selection clipboard -o)" >> ~/.cyrus/.env
```

**Universal fallback:**
```bash
read -s -p "Paste your Anthropic API key: " val && printf 'ANTHROPIC_API_KEY=%s\n' "$val" >> ~/.cyrus/.env && echo " ✓ Saved"
```

### Option 3: Separate OAuth Token

This is for when the user wants to generate a token for a different account than the one currently logged in (e.g., running `claude setup-token` on another machine).

Append the placeholder, then open the file for the user to paste:

```bash
grep -q '^CLAUDE_CODE_OAUTH_TOKEN=' ~/.cyrus/.env || echo 'CLAUDE_CODE_OAUTH_TOKEN=' >> ~/.cyrus/.env
```

```bash
# macOS
code --new-window ~/.cyrus/.env 2>/dev/null || open -a TextEdit ~/.cyrus/.env
# Linux
code ~/.cyrus/.env 2>/dev/null || xdg-open ~/.cyrus/.env
```

Tell the user:

> 1. On the other machine, run `claude setup-token`
> 2. Copy the token it outputs
> 3. I've opened `~/.cyrus/.env` — find `CLAUDE_CODE_OAUTH_TOKEN=` and paste the token after the `=`
> 4. Save and close

### Option 4: Third-Party Provider

Inform the user:

> For third-party providers, you'll need to set provider-specific environment variables.
> See [Third-Party Integrations](https://docs.anthropic.com/en/docs/claude-code/bedrock-vertex) for details.
>
> Common configurations:
>
> **AWS Bedrock:**
> ```
> CLAUDE_CODE_USE_BEDROCK=1
> AWS_REGION=us-east-1
> ```
>
> **Google Vertex AI:**
> ```
> CLAUDE_CODE_USE_VERTEX=1
> CLOUD_ML_REGION=us-east5
> ANTHROPIC_VERTEX_PROJECT_ID=your-project-id
> ```

Guide the user to add the appropriate variables to `~/.cyrus/.env` using the clipboard-to-env pattern.

## Step 4: Verify

After the user runs the command, verify the key was written:

```bash
grep -c -E '^(ANTHROPIC_API_KEY|CLAUDE_CODE_OAUTH_TOKEN|CLAUDE_CODE_USE_BEDROCK|CLAUDE_CODE_USE_VERTEX)=' ~/.cyrus/.env
```

If the count is 0, the credential was not saved. Ask the user to try again.

## Completion

> ✓ Claude Code authentication configured.

Excerpt, first 4,849 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 95804c4

git clone https://github.com/ceedaragents/cyrus
git -C cyrus checkout 95804c41da68

The author documents no install command, so this uses the repository directly. Our verdict covers commit 95804c41da68 only.

Works in which agents?

What its own text says, not a test

AgentStatusMore
Claude CodeMentioned in its text, not testedSkills for Claude Code · About Claude Code
“Mentioned” means the skill’s text names the agent. We have not run it there.

Audit history

Every audit, every source change

  1. ✕BLOCK D 69 / 100

    first audit · source 95804c41da68

Similar skills

Questions about cyrus-setup-claude-auth

What does the cyrus-setup-claude-auth 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-claude-auth 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-claude-auth access on my machine?
The audit observed no filesystem, network or shell use at all in its source.
Which assistants does cyrus-setup-claude-auth work with?
Its documentation mentions claude-code. That is what the text claims, not a compatibility test we ran.
How current is this page?
The grade is for one exact copy of the source (95804c41da68), read on 2026-10-21. 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.