ClawHub
Back to skill

Security audit

Construction Claim Strategy

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed local construction-claim planning aid with no evidence of network access, hidden persistence, exfiltration, or automatic high-impact actions.

Use this as a planning and drafting aid, not as legal advice. Because the outputs can affect claim rights, disclosure posture, deadlines, and dispute strategy, have qualified legal or contract professionals review any real submission before sending it externally.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Memory PoisoningPersistent Context Injection, Context Window Stuffing, Memory Manipulation
  • 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
Findings (5)

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The declared description promises extensive construction-claim strategy functionality, including assessment, argument development, disclosure control, risk analysis, and guidance across multiple standard forms and dispute topics. The actual code chunk contains only a version string and a footer formatter. While versioning/footer utilities can be supporting implementation details, this chunk by itself does not implement or evidence the declared core behavior. Therefore, the description does not accurately represent what this supplied code chunk actually does.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The declared description presents an end-user construction-claims strategy advisor that performs substantive planning and legal/contractual guidance. The supplied code chunk does not implement that advisory functionality. Instead, it is a test file whose primary purpose is QA/security validation of another script. Its behavior is materially different: it executes smoke tests, inspects CLI behavior, reads/writes files for test purposes, and parses source code AST to enforce import restrictions. Those are undeclared and unrelated operational capabilities compared with the declared strategic-planning framework. This is a clear description-behavior mismatch.

Content

No source excerpt is available for this finding.

Memory Manipulation

High
Category
Memory Poisoning
Confidence
80% confidence
Finding

Skill manipulates agent memory, state, or stored context. Memory corruption can alter personality, override safety rules, or cause unpredictable behavior.

Content

Scanner excerpt · references/expert_engagement.md (reported line 42)May include surrounding context.

md
## Expert Reports — What Tribunals Expect

- Clear statement of qualifications and experience
- Independence declaration
- Stated methodology and assumptions
- Clear opinion on disputed issues

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
77% confidence
Finding

This markdown file provides detailed guidance for preparing construction claim responses, including scope control, disclosure control, and argument strategy, which could materially affect legal or commercial outcomes. Although L133 states it is not legal advice, the document does not warn users to seek qualified review before using the framework in real matters, so the safety implications for user interests are only partially disclosed.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.