ClawHub
Back to skill

Security audit

Cursor

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed Cursor workflow guide that emphasizes consent, local scoping, and review before remote or high-trust features.

Install this only if you are comfortable with Cursor workflows potentially sending selected prompts, code context, diffs, and GitHub workflow metadata to Cursor or GitHub services when those features are used. Review and set the ~/cursor/ memory boundaries before enabling Background Agents, Bugbot, remote MCP, or non-interactive cursor-agent runs.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Hidden Instructions

High
Category
Prompt Injection
Confidence
70% confidence
Finding

Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.

Content

Scanner excerpt · memory-template.md (reported line 15)May include surrounding context.

md
integration: pending | done | declined

## Context
<!-- What the user uses Cursor for and why -->
<!-- Example: repo-aware implementation and review with local CLI only, no Background Agents -->

## Activation Boundaries

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 129)May include surrounding context.

md
| https://cursor.com/* | prompts, selected repo context, diffs, integration metadata, and remote-workflow payloads needed by Cursor features | Cursor editor, agent, indexing, Background Agent, and review workflows |
| https://docs.cursor.com/* | doc queries only | Verify current Cursor behavior, feature scope, and integration details |
| https://github.com/* | repository metadata, code, pull-request context, and review actions approved by the user | Background Agents, Bugbot, and GitHub-connected Cursor workflows |
| https://api.github.com/* | repository, branch, PR, and review metadata approved by the user | GitHub API access used by Cursor-connected review and remote execution flows |
| https://{user-approved-mcp-host} | request payloads required by the approved MCP server | Optional MCP tool access beyond the local machine |

No other data is sent externally unless the user explicitly approves more hosts or integrations.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The file asks whether the skill should activate whenever several broad topics such as Cursor, MCP, or related tools 'come up,' and whether it should 'jump in proactively.' This does not define specific trigger phrases or clear exclusion conditions, creating a risk of unintended invocation across ordinary discussion of these topics.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.