Back to skill

Security audit

Openclaw Money Maker

Security checks for vulnerabilities and agentic risk

Overview

The skill is an earnings guide, but it needs review because it asks users to keep multiple service credentials in plaintext and set up recurring agent jobs with broad posting behavior.

Install only if you are comfortable with an agent using earning-platform accounts and network APIs. Do not store real passwords or API keys in the suggested plaintext file unless you first restrict permissions, exclude it from backups and sync, and use limited, revocable tokens where possible. Treat the cron and HEARTBEAT sections as optional and avoid enabling them until posting, claiming, spending, and submission actions require explicit approval.

Vulnerability Patterns
  • System PersistenceInstalls backdoors, hooks, services, or scheduled tasks that survive the run
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Findings (2)

T06 · System Persistence

Error
Location
SKILL.md:227
Finding

Persistent Autonomous Tasks Installed Through Cron and Heartbeat Configuration

Content
View full analysis

Vulnerability Details

File Location: SKILL.md, lines 227-246
Vulnerability Type: Persistent scheduled agent activity
Risk Level: High

Vulnerable Code:

markdown
## Automation Setup

### Cron Jobs

Add to `~/.openclaw/cron/jobs.json`:

```json
{
  "id": "bounty-search",
  "schedule": {"everyMs": 21600000},
  "payload": {
    "kind": "agentTurn",
    "message": "Search for new bounties on GitHub and PayAClaw..."
  }
}

Heartbeat Tasks

Add to HEARTBEAT.md:

markdown
## Money Making Tasks (Every 2 hours)
1. Check PayAClaw for new tasks
2. Search GitHub bounties
3. Check PR/claim status
4. Post content if needed
text

### Technical Analysis

The Skill instructs the user or agent to modify persistent OpenClaw configuration files. The cron entry schedules a new agent turn every six hours, while the heartbeat instructions request activity every two hours. These changes survive the original Skill invocation and can cause recurring network access and agent execution in future sessions.

Persistence is not necessary for the Skill's core purpose as an earning guide. It is an optional automation capability and therefore should not be installed by default. The heartbeat instruction `Post content if needed` is particularly broad: it does not identify the destination, content, authorization boundaries, or conditions under which publication is permitted. If interpreted autonomously, it could result in external side effects without approval for each action.

The supplied cron payload only directs searches and does not explicitly contain a backdoor or destructive command. Nevertheless, installing recurring agent turns expands the Skill's authority beyond the initiating session and violates least-privilege expectations unless the user gives explicit, informed consent.

### Attack Path

1. The user or agent loads the Skill and follows its automation instructions.
2. A recu
...[truncated 1189 chars]
Remediation
View remediation

Remediation Suggestions

  • Remove cron and heartbeat modification from the default or quick-start workflow.
  • Present automation as a clearly optional feature and obtain explicit informed consent before writing persistent state.
  • Display the exact schedule, destinations, anticipated network requests, credentials used, and possible external side effects before installation.
  • Restrict recurring jobs to read-only discovery operations.
  • Require separate, per-action confirmation before submitting work, claiming bounties, creating posts, or publishing content.
  • Replace Post content if needed with narrowly defined conditions and an approval gate.
  • Add resource, rate, and execution limits, along with logging and failure controls.
  • Provide exact commands or instructions for disabling and removing every cron and heartbeat entry created by the Skill.
  • Prefer a foreground, one-time command when persistent execution is not essential.

T09 · Insecure Skill Coding Practices

Warning
Location
SKILL.md:58
Finding

API Keys and Password Stored Together in a Plaintext Credentials File

Content
View full analysis

Vulnerability Details

File Location: SKILL.md, lines 58-77
Vulnerability Type: Insecure plaintext credential storage
Risk Level: Medium

Vulnerable Code:

markdown
### Step 2: Save Your Credentials

Create `~/.config/openclaw-earnings/credentials.json`:

```json
{
  "payaclaw": {
    "agent_id": "...",
    "api_key": "..."
  },
  "openclawlog": {
    "username": "...",
    "password": "...",
    "xmlrpc_url": "https://openclawlog.com/xmlrpc.php"
  },
  "moltbook": {
    "agent_name": "...",
    "api_key": "..."
  }
}
text

### Technical Analysis

The Skill recommends storing reusable API keys and an OpenClawLog password together in an unencrypted JSON file. It does not instruct the user to set restrictive directory and file permissions, use an operating-system credential store, prevent backup or synchronization exposure, redact secrets from diagnostics, or rotate credentials.

A file under the user's home directory is not necessarily public, but its confidentiality depends on the host's default umask, filesystem permissions, backup configuration, and access granted to other applications or agent tools. Consolidating credentials for three services also increases the effect of a single local disclosure.

Credential storage is relevant to authenticated platform operations, but plaintext storage without access-control guidance exceeds the minimum exposure necessary. The Skill does not contain code that transmits this file or sends credentials to an undeclared destination; exploitation requires another process, user, backup reader, or tool to gain local read access.

### Attack Path

1. The user follows the Skill and creates `credentials.json` with live platform credentials.
2. The file receives default permissions that may be broader than intended, or it is included in backups, synchronization, diagnostics, or agent-readable filesystem scope.
3. A local process, another user
...[truncated 804 chars]
Remediation
View remediation

Remediation Suggestions

  • Store secrets in an operating-system keychain or a dedicated secret manager instead of a plaintext JSON document.
  • Prefer narrowly scoped, revocable API tokens over account passwords.
  • Use separate credentials for each service and grant only permissions required by the documented workflow.
  • If file storage is unavoidable, create the directory with mode 0700 and the credential file with mode 0600.
  • Create the file atomically with restrictive permissions rather than creating it broadly and correcting permissions afterward.
  • Exclude the credential path from source control, cloud synchronization, backups, logs, diagnostics, and support bundles where feasible.
  • Never print credential values in commands, error messages, or agent transcripts.
  • Document credential rotation and revocation procedures.
  • Consider separate files or secret entries per service to reduce the exposure caused by compromise of one storage location.
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Rogue AgentSelf-Modification, Session Persistence
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (5)

Credential Access

High
Category
Privilege Escalation
Confidence
94% confidence
Finding

The skill instructs storing sensitive credentials, including API keys and a plaintext password, in a predictable filesystem location. That creates a concentrated credential target for malware, other local users, misconfigured sync/backups, or accidental disclosure, and could enable account takeover across multiple services.

Content

Scanner excerpt · SKILL.md (reported line 60)May include surrounding context.

Step 2: Save Your Credentials

Create ~/.config/openclaw-earnings/credentials.json:

json
{

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 43)May include surrounding context.

bash
# PayAClaw - Task platform
curl -X POST https://payaclaw.com/api/agents/register \
  -H "Content-Type: application/json" \
  -d '{"agent_name":"YourAgentName","description":"Your capabilities","capabilities":["coding","writing"]}'

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill explicitly instructs users to store multiple platform credentials, including API keys and a plaintext password, in a local JSON file without any guidance on file permissions, encryption, or secret management. This increases the chance of credential exposure through local compromise, backups, logs, or accidental publication.

Content

No source excerpt is available for this finding.

Session Persistence

Medium
Category
Rogue Agent
Confidence
88% confidence
Finding

Persisting long-lived session material and API credentials in a static local file increases the window of compromise and makes unauthorized reuse easier after device compromise or accidental exposure. In this skill, persistence spans several external platforms, so one leaked file could enable durable misuse of multiple accounts.

Content

Scanner excerpt · SKILL.md (reported line 60)May include surrounding context.

Step 2: Save Your Credentials

Create ~/.config/openclaw-earnings/credentials.json:

json
{

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
84% confidence
Finding

The automation section directs users to create recurring jobs and heartbeat tasks that will repeatedly query external services and potentially post or interact with platforms, but it omits warnings about continuous network activity, rate limits, account actions, and unintended spend or abuse. In an autonomous agent context, unsupervised recurring external actions materially raise operational and security risk.

Content

No source excerpt is available for this finding.