ClawHub
Back to skill

Security audit

Freelance Pipeline Automation

Security checks for vulnerabilities and agentic risk

Overview

This skill is a local freelance lead tracker with overstated documentation, but I found no hidden network access, credential theft, persistence, or destructive behavior.

Install only if you want a simple local JSON-based lead tracker. Do not rely on it for automated marketplace discovery, browser automation, true recency scoring, or proposal generation unless those features are added or documented accurately. Review the configured profile and pipeline paths before use because the script will read and write those local files.

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 (6)

Tp4

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding

The skill documentation materially overstates what the tool does, claiming automated job discovery, specific weighted scoring, proposal workflow integration, and browser/gated-site support that are not actually implemented. This is dangerous because an agent or user may rely on nonexistent automation and inaccurate scoring/business logic, leading to unsafe assumptions, bad decisions, and misuse of the skill in production workflows.

Content

No source excerpt is available for this finding.

Lp1

High
Category
MCP Least Privilege
Confidence
75% confidence
Finding

The skill uses 'env' capability that is not listed in its permissions. This may indicate deceptive intent or missing permission declarations.

Content

No source excerpt is available for this finding.

Lp1

High
Category
MCP Least Privilege
Confidence
75% confidence
Finding

The skill uses 'file_write' capability that is not listed in its permissions. This may indicate deceptive intent or missing permission declarations.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The manifest specifies lead scoring as skill match (50%) + budget (30%) + recency (20%). The code starts from a fixed base score of 50, adds budget bonuses and skill-match bonuses, and never accepts or calculates recency, so the implemented scoring semantics differ materially from the stated behavior.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
98% confidence
Finding

The manifest describes discovering jobs on Upwork, Fiverr, LinkedIn, and niche boards, generating personalized proposals, and tracking the pipeline through payment. In the actual CLI implementation, commands only add leads manually, score them, print digests, update status, and list records; there is no job discovery workflow and no command that generates or stores proposals despite the presence of a proposal helper.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Low
Category
Not specified by scanner
Confidence
92% confidence
Finding

Although a generate_proposal function exists, the CLI never calls it and there is no command to produce, save, or display proposals. This means the advertised proposal-generation capability is not part of the actual user-facing behavior of the script.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.