Back to skill

Security audit

trading-book-library

Security checks for vulnerabilities and agentic risk

Overview

This skill is a static trading-book reference, but users should be aware it points agents to a hardcoded local text corpus outside the packaged skill.

Install only if you are comfortable with an agent being told to read a specific local text corpus for chapter lookup. Expect the full-text lookup to fail unless that exact path exists, and verify rights/provenance before using or redistributing extracted book text.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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 (2)

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill discloses a concrete local filesystem path and explicitly instructs the agent to use file-search and file-read operations against that path. That expands the skill from a passive knowledge summary into a capability that can access host-local data, creating risk of unintended data exposure, path probing, and retrieval of content outside the declared skill boundary if the referenced workspace is broader than expected.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill explicitly instructs the agent to access a local temp-directory file containing a large extracted corpus via search_files and read_file, but provides no guardrails about provenance, sensitivity, or copyright constraints. Even though the path is specific, normalizing agent behavior around reading local filesystem content can expose unintended local data and encourage retrieval of material that may be proprietary or sensitive.

Content

No source excerpt is available for this finding.