ClawHub
Back to skill

Security audit

Context Near Overflow

Security checks for vulnerabilities and agentic risk

Overview

This is a short instructional skill about managing near-full context windows and does not request access, execute code, or persist data.

This skill appears safe to install as general workflow guidance. Users should treat it as advice for managing long conversations, not as a tool that performs actions or changes system settings.

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
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

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 · SKILL.md (reported line 12)May include surrounding context.

md
# context-near-overflow

When a conversation or task grows large enough to fill the context window, the model begins silently dropping earlier content. The output doesn't error — it degrades. The model appears to be working but is operating on a truncated view of the task, producing answers that are incomplete, inconsistent, or contradictory to earlier parts of the session.

## Symptoms

Static analysis

No suspicious patterns detected.