ClawHub
Back to skill

Security audit

QA & Testing Engine

Security checks for vulnerabilities and agentic risk

Overview

This is a documentation-only QA/testing skill whose flagged phrases are testing examples rather than hidden access or execution behavior.

Install only if you want your agent to use a broad QA and testing playbook. Review security-test payloads before running them against real systems, and keep production testing limited to explicitly approved smoke or staging-safe checks.

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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Memory PoisoningPersistent Context Injection, Context Window Stuffing, Memory Manipulation
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (4)

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 · SKILL.md (reported line 278)May include surrounding context.

md
| Symptom | Likely Cause | Fix |
|---|---|---|
| Passes locally, fails in CI | Timing/race condition | Add explicit waits, check CI resource limits |
| Fails intermittently | Shared state between tests | Isolate test data, reset state |
| Fails after deploy | Environment difference | Check env vars, API versions, feature flags |
| Fails at specific time | Time-dependent logic | Mock dates/times, avoid time-sensitive assertions |
| Fails in parallel | Resource contention | Use unique ports/DBs per worker |

Credential Access

High
Category
Privilege Escalation
Confidence
60% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · SKILL.md (reported line 417)May include surrounding context.

md
injection_payloads:
  sql: ["' OR 1=1--", "'; DROP TABLE users;--", "1 UNION SELECT * FROM users"]
  xss: ["<script>alert(1)</script>", "<img onerror=alert(1) src=x>", "javascript:alert(1)"]
  path_traversal: ["../../etc/passwd", "..\\..\\windows\\system32", "%2e%2e%2f"]
  command: ["; ls -la", "| cat /etc/passwd", "$(whoami)", "`id`"]
  
boundary_values:

Credential Access

High
Category
Privilege Escalation
Confidence
60% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · SKILL.md (reported line 418)May include surrounding context.

md
injection_payloads:
  sql: ["' OR 1=1--", "'; DROP TABLE users;--", "1 UNION SELECT * FROM users"]
  xss: ["<script>alert(1)</script>", "<img onerror=alert(1) src=x>", "javascript:alert(1)"]
  path_traversal: ["../../etc/passwd", "..\\..\\windows\\system32", "%2e%2e%2f"]
  command: ["; ls -la", "| cat /etc/passwd", "$(whoami)", "`id`"]
  
boundary_values:

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The skill exposes very generic natural-language trigger phrases such as requests to create strategies, write tests, review tests, or assess test health. Because these overlap with ordinary developer conversation, an agent may invoke the skill unintentionally, causing prompt hijacking of workflow, unexpected tool behavior, or disclosure of user/project context to the skill when the user did not explicitly request it.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.