Back to skill

Security audit

teletype-support

Security checks for vulnerabilities and agentic risk

Overview

This skill gives clear Teletype support workflows and uses customer-message tools in a disclosed, approval-gated way.

Install only if you intend to connect an agent to a Teletype support project. Store the API token in private configuration or environment variables, verify the project identity with get_capabilities, and review every outbound reply or conversation-changing action before approving it.

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 (1)

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 · references/installation.md (reported line 5)May include surrounding context.

md
This skill describes support workflows. It requires a working connection to the Teletype MCP server and a Teletype project with Public API access. The skill itself does not authenticate or install a server.

The hosted endpoint is `https://mcp.teletype.app/mcp` with Streamable HTTP transport. Supply the project's Public API token in `X-Teletype-Api-Token` through your client's private settings or secret storage. Do not ask the user to paste the token into a chat, skill file, public repository, or installation link.

Use the [client setup guide](https://github.com/Teletype-App/teletype-mcp-server/blob/main/docs/CLIENTS.md) or its [Russian version](https://github.com/Teletype-App/teletype-mcp-server/blob/main/docs/ru/CLIENTS.md). They include Cursor, Claude Code, Codex, Gemini CLI, and OpenClaw configurations.

Static analysis

No suspicious patterns detected.