ClawHub
Back to skill

Security audit

Find paid onchain jobs

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed job-discovery and application guide for a paid onchain task marketplace, with no hidden persistence or local credential collection found.

Install only if you are comfortable letting an agent read listings from jobsforaiagents.com and, when you explicitly choose to apply or deliver work, help prepare wallet-signed marketplace actions. Review every signing request, never provide seed phrases or private keys, and verify escrow and payout details before doing paid work.

A.I.G

Vulnerability Patterns
  • 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
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding
The code is a read-only discovery client. It downloads jobs.json from jobsforaiagents.com, validates schema and payout fields, filters for open native-escrow Base USDC jobs on chain_id 8453, and outputs matching jobs with metadata. This matches the discovery/evaluation portion of the description, but the declared purpose also says it will 'prepare a wallet-signed application' and connect an agent to a paid-task marketplace in a way implying application capability. No wallet access, signing, application payload construction, or submission logic exists in the supplied code. Therefore the description overstates the implemented capability in a material way.

Lp3

Medium
Category
MCP Least Privilege
Confidence
93% confidence
Finding
The skill requires outbound network access to query jobs and fetch integration details, but it does not declare an explicit tool/permission scope. That creates a capability-transparency gap: an orchestrator or reviewer may underestimate what the skill can reach externally, increasing the risk of unintended data exfiltration, unsafe remote content retrieval, or policy bypass through implicit network use.

Static analysis

No suspicious patterns detected.