Back to skill

Security audit

Paper Reading Method Inspiration

Security checks for vulnerabilities and agentic risk

Overview

This skill is a structured research-workflow guide that reads and writes local research notes, with confirmation gates and no hidden code or credential handling.

Before installing, expect the skill to create and update research notes under the configured workspace and to search for papers unless you opt out or search is unavailable. Use it on a project workspace where those Markdown artifacts are appropriate, and review the generated candidate methods because they are explicitly provisional inspiration, not a final method design.

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
  • 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)

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 322)May include surrounding context.

md
- source URLs for every source paper used
- risks and weak links
- light downstream routing hints
- a clear statement that the library does not contain a Committed Method Design

Write or update `method_candidate_libraries.md` as a cross-problem summary.