Back to skill

Security audit

euthyna (English)

Security checks for vulnerabilities and agentic risk

Overview

ClawScan marked this skill suspicious because its guidance includes running untrusted project tests and exploit demonstrations without requiring safeguards; the scan is partial.

Treat this as a review-needed skill. Before installing or following its audit steps, consider whether you can run tests and exploit demonstrations only in an isolated, disposable environment; the scan was partial because a referenced local command could not be resolved.

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
Vulnerability Patterns
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Memory PoisoningPersistent Context Injection, Context Window Stuffing, Memory Manipulation
  • 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 (3)

Context Window Stuffing

Medium
Category
Memory Poisoning
Confidence
85% confidence
Finding

Skill attempts to fill the context window with filler content, displacing legitimate instructions and safety constraints. This can degrade agent performance or bypass safety boundaries.

Content

Scanner excerpt · references/bug-classes.md (reported line 20)May include surrounding context.

md
| **logic defect** | default to **not a false positive**. "Logic defects pass every boundary check" — **do not let clean static analysis convince you it is a false positive** |
| remaining classes | see below, class by class |

Flip these two rows and you flood memory corruption with false positives while missing real logic-defect vulnerabilities.

---

Coverage-generation instructions run audited project code without warning about executing untrusted code.

Medium
Category
Output Handling
Confidence
95% confidence
Finding

The coverage instructions direct the agent to execute audited-project tests, including Go tests, pytest, and cargo test, without requiring isolation or establishing trust in the project. In a security-audit context, repository-controlled tests, plugins, and build hooks may be hostile and could execute with the agent's filesystem, credential, and network access.

Content

No source excerpt is available for this finding.

The workflow requires running exploit PoCs without warning about their effects on user data or system integrity.

Medium
Category
Output Handling
Confidence
96% confidence
Finding

The workflow calls for executable exploit demonstrations and real output without requiring a disposable target, containment, or approval for destructive effects. Exploit PoCs can intentionally trigger command execution, data corruption, disclosure, or service failure; restricting which subagent writes the PoC does not contain its execution.

Content

No source excerpt is available for this finding.