ClawHub
Back to skill

Security audit

test-updates

Security checks for vulnerabilities and agentic risk

Overview

The skill is a coherent test-generation helper, but its test execution and in-place mutation-testing guidance can alter project files without enough safeguards.

Install only if you are comfortable with a testing skill that may run project code and modify test or source files. Use a virtual environment, review generated changes before committing, and avoid mutation testing in your main working tree unless you have backups or a disposable worktree.

A.I.G

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)

T08 · Insecure Dependencies

Warning
Location
SKILL.md:76
Finding

Unpinned Third-Party Dependency Installation

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
modules/quality-validation.md:218
Finding

Mutation Testing Can Leave Source Files Corrupted After Interrupted Execution

Content
View full analysis
Remediation
View remediation

SkillSpector

Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (5)

Anti-Refusal Statement

High
Category
Anti-Refusal
Confidence
80% confidence
Finding

Skill instructs the agent to omit warnings, disclaimers, or ethical commentary. Stripping safety caveats hides risk from the user and is a common jailbreak preamble.

Content

Scanner excerpt · modules/tdd-workflow.md (reported line 72)May include surrounding context.

md
### Verification Steps
1. **Run the test**: Must pass
2. **Check other tests**: All still passing
3. **No warnings/errors**: Clean execution

### REFACTOR Phase: Clean Up

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The trigger phrase "testing" is extremely broad for an automatically invokable skill. In normal development conversations, this can cause the skill to activate in many unrelated contexts, increasing the chance that it influences workflows, runs analysis/generation steps, or steers the agent unexpectedly when the user did not explicitly request it.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The trigger phrase "code changes require new or updated test coverage" is ambiguous and policy-like rather than a precise invocation cue. Because it overlaps with common review language, the skill may activate based on routine discussion of code quality rather than an explicit request, leading to unintended execution paths or agent behavior changes.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

This module explicitly describes executing tests via pytest and performing mutation testing by applying changes to source files and restoring them, but it provides no safety constraints, user confirmation requirements, sandboxing guidance, or warnings about code execution and file modification risks. In an agent skill, that is dangerous because tests and project code can contain attacker-controlled behavior, so blindly running them or mutating workspace files can trigger arbitrary code execution, side effects, or repository corruption.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% 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 · modules/test-generation.md (reported line 371)May include surrounding context.

md
Then it must NOT find manipulative imperatives.

    Imperative language causes Claude to ignore user intent
    and force actions without consent.
    """
    forbidden = [
        "YOU MUST EXECUTE THIS NOW",

Static analysis

No suspicious patterns detected.