Back to skill

Security audit

InboxPip

Security checks for vulnerabilities and agentic risk

Overview

ClawScan judged this skill benign with high confidence; scanners found no AIG issues, while noting that sending approved email may go beyond the described inbox and draft tasks.

The verdict is benign, not risk-free. Before installing, consider whether you want a skill whose reported workflow can send an email after approval, beyond preparing drafts. Scanning was complete for the three reported components, though code-specific behavioral analyzers were not applicable.

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
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
80% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 19)May include surrounding context.

md
Discover the host's available MCP tools. InboxPip's canonical endpoint is `https://inboxpip.com/mcp`; the connection may be named `applemailmcp` and the host may prefix tool names. Use the actual discovered schemas. If no connection exists, give the user one step: open [Connect your AI](https://inboxpip.com/connect), connect their own mailbox, and configure InboxPip through their host's supported MCP connection settings. Stop until the tools are available. Do not invent tool names or claim the skill configures MCP for them.

Credentials belong in InboxPip's account setup and the host's approved connection flow. Never ask the user to paste a mailbox password, API key or OAuth token into chat. Do not read browser profiles, local token files or another agent's credentials. This skill contains no executable, installer, background job or telemetry.

## Choose the inbox first

The workflow authorizes sending email although the manifest describes only inbox triage, message lookup, and draft preparation.

Medium
Category
Excessive Agency
Confidence
94% confidence
Finding

The manifest at L03 describes preparing reply drafts, but L45 instructs the agent to call send_message after obtaining approval. Sending email is an externally visible action beyond draft preparation; the approval requirement reduces risk but does not resolve the mismatch in declared scope.

Content

No source excerpt is available for this finding.

Scope Creep

Low
Category
Excessive Agency
Confidence
70% confidence
Finding

Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.

Content

Scanner excerpt · LICENSE (reported line 7)May include surrounding context.

text
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.