Otp Challenger

Enable agents and skills to challenge users for fresh two-factor authentication proof (TOTP or YubiKey) before executing sensitive actions. Use this for identity verification in approval workflows - deploy commands, financial operations, data access, admin operations, and change control.

Security Scan

The skill's code and requirements largely match an OTP challenger, but there are concerning elements (an optional failure hook that can run arbitrary shell commands, examples that encourage destructive hooks, and minor metadata inconsistencies) that warrant cautious review before installing.

Purpose & Capability

Name/description (TOTP + YubiKey verification) align with the included scripts (verify.sh, check-status.sh, generate-secret.sh), the declared binaries (openssl, curl, base64, jq, python3, optionally oathtool) are reasonable for the stated functionality, and the conditional env vars (OTP_SECRET, YUBIKEY_CLIENT_ID, YUBIKEY_SECRET_KEY) are the exact credentials needed.

Instruction Scope

Runtime instructions and scripts operate within the expected scope: they read config (~/.openclaw/config.yaml) or env vars for secrets, maintain a local state file (memory/otp-state.json), call Yubico's API for YubiKey validation, and log audit events. They do not appear to read unrelated system files. However the SKILL.md and scripts document an OTP_FAILURE_HOOK that the skill will execute on failure events; that hook runs arbitrary shell commands and is therefore out-of-band behavior that increases risk if misconfigured or abused.

Install Mechanism

Install spec only references Homebrew formulas (jq, python3, oath-toolkit). No downloads from untrusted URLs, no archive extraction, and the repo files are present in the skill bundle. Homebrew usage is proportionate for these native binaries.

Credentials

No required global credentials are demanded by default; required credentials are conditional and match their purpose (TOTP secret or Yubico client id/secret). However OTP_FAILURE_HOOK is an environment/config option that allows arbitrary commands to run as the agent user on verification failures. The README even includes an example hook that kills OpenClaw — this is a high-risk capability that is disproportionate unless the operator explicitly intends it and has locked down the hook's contents and permissions.

Persistence & Privilege

The skill does not request forced or persistent platform-level privileges (always:false). It writes state to its own workspace memory file and may write audit logs, which is expected. The main persistence/privilege risk is the configurable failure hook and any scripts it calls; those run with the same privileges as the agent and can have system impact if misused.

What to consider before installing

This skill appears to implement what it claims, but review these items before installing: 1) Inspect verify.sh (and related scripts) yourself to confirm there are no unexpected network endpoints beyond api.yubico.com and no hidden behavior. 2) Don’t set OTP_FAILURE_HOOK to an unrestricted or destructive script; prefer a safe notifier (append-only logging or an alert sender) and lock its file permissions. 3) Keep OTP_SECRET and YUBIKEY_SECRET_KEY in a secure secret manager (1Password/Bitwarden) and avoid plaintext in ~/.openclaw/config.yaml when possible. 4) Verify the skill source — SKILL.md references a GitHub repo, but the registry metadata shows no homepage; prefer installing from a verified upstream repository (and check commit history/signature). 5) Run the scripts in a controlled environment or sandbox first, and review audit logs after test runs. If you rely on autonomous agent invocation, be extra cautious about enabling any hook that executes commands.

OTP Identity Challenge Skill

Challenge users for fresh two-factor authentication before sensitive actions.

When to Use

Require OTP verification before:

• Deploy commands (kubectl apply, terraform apply)
• Financial operations (transfers, payment approvals)
• Data access (PII exports, customer data)
• Admin operations (user modifications, permission changes)

Scripts

verify.sh

Verify a user's OTP code and record verification state.

./verify.sh <user_id> <code>

Parameters:

• user_id - Identifier for the user (e.g., email, username)
• code - Either 6-digit TOTP or 44-character YubiKey OTP

Exit codes:

• 0 - Verification successful
• 1 - Invalid code or rate limited
• 2 - Configuration error (missing secret, invalid format)

Output on success:

✅ OTP verified for <user_id> (valid for 24 hours)
✅ YubiKey verified for <user_id> (valid for 24 hours)

Output on failure:

❌ Invalid OTP code
❌ Too many attempts. Try again in X minutes.
❌ Invalid code format. Expected 6-digit TOTP or 44-character YubiKey OTP.

check-status.sh

Check if a user's verification is still valid.

./check-status.sh <user_id>

Exit codes:

• 0 - User has valid (non-expired) verification
• 1 - User not verified or verification expired

Output:

✅ Valid for 23 more hours
⚠️ Expired 2 hours ago
❌ Never verified

generate-secret.sh

Generate a new TOTP secret with QR code (requires qrencode to be installed).

./generate-secret.sh <account_name>

Usage Pattern

#!/bin/bash
source ../otp/verify.sh

if ! verify_otp "$USER_ID" "$OTP_CODE"; then
  echo "🔒 This action requires OTP verification"
  exit 1
fi

# Proceed with sensitive action

Configuration

Required for TOTP:

• OTP_SECRET - Base32 TOTP secret

Required for YubiKey:

• YUBIKEY_CLIENT_ID - Yubico API client ID
• YUBIKEY_SECRET_KEY - Yubico API secret key (base64)

Optional:

• OTP_INTERVAL_HOURS - Verification expiry (default: 24)
• OTP_MAX_FAILURES - Failed attempts before rate limiting (default: 3)
• OTP_STATE_FILE - State file path (default: memory/otp-state.json)

Configuration can be set via environment variables or in ~/.openclaw/config.yaml:

security:
  otp:
    secret: "BASE32_SECRET"
  yubikey:
    clientId: "12345"
    secretKey: "base64secret"

Code Format Detection

The script auto-detects code type:

• 6 digits (123456) → TOTP validation
• 44 ModHex characters (cccccc...) → YubiKey validation

ModHex alphabet: cbdefghijklnrtuv

State File

Verification state stored in memory/otp-state.json. Contains only timestamps, no secrets.

Human Documentation

See README.md for:

• Installation instructions
• Setup guides (TOTP and YubiKey)
• Security considerations
• Troubleshooting
• Examples