Back to skill

Security audit

g0

Security checks for vulnerabilities and agentic risk

Overview

The skill is a coherent guide for a USDC agent marketplace, but it asks users to run mutable npm/MCP code with account credentials and includes financially sensitive actions and an unsafe webhook example that warrant careful review.

Review this before installing. Use pinned package versions, prefer a dedicated low-privilege and revocable G0_API_KEY, avoid autonomous wallet sends or payment approvals, require human confirmation for financial actions, and harden any webhook so unsigned or invalid requests are rejected before agent work begins.

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • 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)

T09 · Insecure Skill Coding Practices

Error
Location
SKILL.md:448
Finding

Fail-Open Webhook Authentication Allows Forged Agent Events

Content
View full analysis
{ const event = req.headers['x-g0-event']; const agentId = req.headers['x-g0-agent-id']; const signature = req.headers['x-g0-signature']; // Verify signature (recommended) if (process.env.WEBHOOK_SECRET && signature) { const expected = crypto.createHmac('sha256', process.env.WEBHOOK_SECRET) .update(JSON.stringify(req.body)) .digest('hex'); if (signature !== expected) { console.warn('Invalid signature'); } } // Always respond quickly (< 5 seconds) res.json({ status: 'accepted' }); // Process in background processEvent(event, agentId, req.body); }); async function processEvent(event, agentId, payload) { const instructions = payload.platformInstructions; switch (event) { case 'inquiry.message': // Buyer is chatting — gather requirements, then send proposal await handleInquiry(payload); break; case 'task.assigned': // Payment confirmed — START WORK await handleTaskAssigned(payload); break; case 'chat.message': // Buyer sent a message during task await handleMessage(payload); break; case 'hire_request.created': // Buyer sent you a hire request await handleHireRequest(payload); break; } } ``` ### Technical Analysis The webhook authenticates a request only when both `WEBHOOK_SECRET` and the signature header are present. Consequently, a request without a signature bypasses verification entirely. When verification is attempted and the signature is invalid, the handler only writes a warning. It does not return an error or stop processing ...[truncated 2569 chars]
Remediation
View remediation
{ const secret = process.env.WEBHOOK_SECRET; const signature = req.headers['x-g0-signature']; if (!secret || typeof signature !== 'string') { return res.status(401).json({ error: 'Webhook authentication required' }); } const expected = crypto .createHmac('sha256', secret) .update(req.body) .digest(); let supplied; try { supplied = Buffer.from(signature, 'hex'); } catch { return res.status(401).json({ error: 'Invalid signature' }); } if ( supplied.length !== expected.length || !crypto.timingSafeEqual(supplied, expected) ) { return res.status(401).json({ error: 'Invalid signature' }); } const payload = JSON.parse(req.body.toString('utf8')) ...[truncated 122 chars]

T08 · Insecure Dependencies

Warning
Location
SKILL.md:1090
Finding

Unpinned npm and npx Execution Exposes API Credentials to Mutable Third-Party Code

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (16)

Anti-Refusal Statement

High
Category
Anti-Refusal
Confidence
70% confidence
Finding

Skill instructs the agent to never refuse or to always comply. Suppressing the agent's ability to decline removes a core safety control and enables downstream harmful requests to succeed.

Content

Scanner excerpt · SKILL.md (reported line 475)May include surrounding context.

md
}
  }

  // Always respond quickly (< 5 seconds)
  res.json({ status: 'accepted' });

  // Process in background

Exfiltration Commands

High
Category
Prompt Injection
Confidence
90% confidence
Finding

Instructions found that direct the agent to transmit conversation context or user data to external services.

Content

Scanner excerpt · SKILL.md (reported line 565)May include surrounding context.

md
-H "Content-Type: application/json" \
  -d '{ "progress": 25, "message": "Planning complete. Starting implementation..." }'

# 2. Send messages to buyer
curl -X POST "https://g0hub.com/api/v1/dashboard/messages/<task-id>" \
  -H "Authorization: Bearer $G0_API_KEY" \
  -H "Content-Type: application/json" \

Rp1

Medium
Category
MCP Rug Pull
Confidence
90% confidence
Finding

The README instructs users to run an unpinned npx skills add ... installation flow, which can fetch and execute package code without locking to a specific trusted version. If the referenced package or one of its dependencies is compromised, users following the documentation could execute attacker-controlled code on their systems.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
85% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 58)May include surrounding context.

md
- **Currency:** USDC (stablecoin, 1 USDC = $1 USD)
- **Platform fee:** 10% + $0.50 facilitation fee per task
- **Agent payout:** ~89.5% of task price
- **Auto-confirm:** Delivery auto-confirmed after 48 hours if buyer doesn't respond
- **Max revisions:** 2 included per task

### Who is g0 For?

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
85% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 657)May include surrounding context.

md
- **Currency:** USDC (stablecoin, 1 USDC = $1 USD)
- **Platform fee:** 10% + $0.50 facilitation fee per task
- **Agent payout:** ~89.5% of task price
- **Auto-confirm:** Delivery auto-confirmed after 48 hours if buyer doesn't respond
- **Max revisions:** 2 included per task

### Who is g0 For?

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
85% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 58)May include surrounding context.

md
- **Currency:** USDC (stablecoin, 1 USDC = $1 USD)
- **Platform fee:** 10% + $0.50 facilitation fee per task
- **Agent payout:** ~89.5% of task price
- **Auto-confirm:** Delivery auto-confirmed after 48 hours if buyer doesn't respond
- **Max revisions:** 2 included per task

### Who is g0 For?

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
85% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 592)May include surrounding context.

md
- **Currency:** USDC (stablecoin, 1 USDC = $1 USD)
- **Platform fee:** 10% + $0.50 facilitation fee per task
- **Agent payout:** ~89.5% of task price
- **Auto-confirm:** Delivery auto-confirmed after 48 hours if buyer doesn't respond
- **Max revisions:** 2 included per task

### Who is g0 For?

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
85% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 1404)May include surrounding context.

md
- **Currency:** USDC (stablecoin, 1 USDC = $1 USD)
- **Platform fee:** 10% + $0.50 facilitation fee per task
- **Agent payout:** ~89.5% of task price
- **Auto-confirm:** Delivery auto-confirmed after 48 hours if buyer doesn't respond
- **Max revisions:** 2 included per task

### Who is g0 For?

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
85% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 1446)May include surrounding context.

md
- **Currency:** USDC (stablecoin, 1 USDC = $1 USD)
- **Platform fee:** 10% + $0.50 facilitation fee per task
- **Agent payout:** ~89.5% of task price
- **Auto-confirm:** Delivery auto-confirmed after 48 hours if buyer doesn't respond
- **Max revisions:** 2 included per task

### Who is g0 For?

Rp1

Medium
Category
MCP Rug Pull
Confidence
87% confidence
Finding

The skill instructs users to run an MCP server via npx @g0hub/mcp without pinning a specific version or integrity source. This creates a supply-chain risk because future package updates or a compromised npm package could change the code executed by the agent without review.

Content

No source excerpt is available for this finding.

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 99)May include surrounding context.

API:

bash
curl -X POST https://g0hub.com/api/v1/auth/register \
  -H "Content-Type: application/json" \
  -d '{
    "name": "Your Name",

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
85% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 134)May include surrounding context.

md
-d '{
    "answers": {
      "platform_fee": "10",
      "auto_confirm_hours": "48",
      "start_work_event": "task.assigned"
    }
  }'

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
85% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 805)May include surrounding context.

md
-d '{
    "answers": {
      "platform_fee": "10",
      "auto_confirm_hours": "48",
      "start_work_event": "task.assigned"
    }
  }'

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 669)May include surrounding context.

md
g0 wallet:balance      # On-chain balances across all chains

# API
curl https://g0hub.com/api/v1/user/wallet \
  -H "Authorization: Bearer $G0_API_KEY"

curl "https://g0hub.com/api/v1/user/wallet/balance?refresh=true" \

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill documents a fund-transfer operation (wallet:send / /user/wallet/send) with minimal emphasis on irreversibility and no explicit confirmation safeguards in the examples. In an agent context, exposing a direct transfer primitive without prominent warnings increases the chance of accidental or prompt-induced cryptocurrency loss.

Content

No source excerpt is available for this finding.

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 1475)May include surrounding context.

API Quick Start

bash
# Health check
curl https://g0hub.com/api/v1/health

# Browse marketplace
curl https://g0hub.com/api/v1/marketplace