Back to skill

Security audit

nitrosend-setup

Security checks for vulnerabilities and agentic risk

Overview

The skill provides coherent, user-guided Nitrosend setup with expected account changes, logo uploads, and optional recurring reports.

Use this for Nitrosend setup, restrict connection access to the intended account or brand, and provide only logos you are comfortable uploading. Review sender defaults and enable recurring reports only if wanted.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

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 20)May include surrounding context.

md
If the user is in Claude Desktop, Claude.ai, Claude Cowork, or this Nitrosend
Claude plugin, connect Nitrosend through the remote MCP connector at
`https://api.nitrosend.com/mcp`. Do not recommend a terminal CLI or API-key
setup for standard Claude Desktop onboarding.

In any other agent (Antigravity, Cursor, Codex, Gemini CLI, VS Code and

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 44)May include surrounding context.

md
If not connected, tell the user:
1. Open the MCP or connectors settings in their client
2. Select the bundled `nitrosend` server if a Nitrosend plugin is installed
   (Claude or Antigravity). Otherwise add `https://api.nitrosend.com/mcp`
   using the guide for that client at
   https://docs.nitrosend.com/integrations/overview
3. Complete the browser sign-in and approval flow

The local-image workflow instructs uploading user data without a privacy disclosure.

Medium
Category
Output Handling
Confidence
88% confidence
Finding

The skill instructs the agent to use nitro_ingest on a local image and then pass its returned media URL into the brand setup. It does not tell the user that this transfers their local image to Nitrosend, leaving an external data transfer undisclosed.

Content

No source excerpt is available for this finding.

Activation phrases include unqualified "onboard" and "configure email," which extend beyond Nitrosend setup.

Low
Category
Trigger Abuse
Confidence
96% confidence
Finding

"Onboard" can refer to unrelated onboarding workflows, and "configure email" can mean setting up an email client or another provider. These phrases are not explicitly restricted to an established Nitrosend context, so they could cause unintended activation.

Content

No source excerpt is available for this finding.