ClawHub
Back to skill

Security audit

box-integration-best-practices

Security checks for vulnerabilities and agentic risk

Overview

This skill is a short Box API best-practices guide with no executable code, install hooks, persistence, or hidden data access.

Before installing, treat this as operational guidance for Box API work. Review any implementation you build from it carefully because Box copy, move, update, upload, webhook, and token-handling workflows can affect real business content, but the skill itself does not execute those actions.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

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

Content

Scanner excerpt · SKILL.md (reported line 14)May include surrounding context.

md
For reviews, trace Box requests through loops and retry paths before changing concurrency or seeking more capacity.

- Reuse cached access tokens. Request fields needed for a decision in the list response when supported, and reuse returned data instead of fetching each item again. `fields` reduces call count only when it replaces a follow-up read.
- Use the largest useful supported page size. For large or changing collections, use [marker pagination](https://developer.box.com/guides/api-calls/pagination/marker-based) where supported; follow `next_marker` to completion and do not treat a marker as a durable checkpoint.
- Use webhooks or events with a catch-up path instead of repeated full scans. Deduplicate deliveries; checkpoint event stream positions after successful processing. Query supported [enterprise metadata templates](https://developer.box.com/guides/metadata/queries/limitations) instead of fetching metadata for every file.
- When content stays in Box, prefer Box-side copy, move, or update operations over download and re-upload.

Static analysis

No suspicious patterns detected.