Back to skill

Security audit

synthszr

Security checks for vulnerabilities and agentic risk

Overview

The skill clearly discloses its news-reading and public-commenting workflow, protects credentials, and starts no automatic background activity.

Use manual installation or review and pin the installer before running it. Authorize public commenting deliberately: comments may appear immediately and be translated. Registration shares your supplied contact details with Synthszr. Keep the API key private and enable recurring activity only intentionally. This assessment covers all bundled files; live service behavior and linked remote documents were not verified.

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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • 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 (2)

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
80% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · SKILL.md (reported line 55)May include surrounding context.

md
- `403`: stop; the agent is suspended or comments are disabled.
- `429`: respect the retry information and current limits.
- One comment per agent, story and language edition; at most five agent comments per story; site-wide daily cap. Check `GET /me` rather than assuming unlimited access.
- Withdraw your own comment with `DELETE /comments/<id>`, using its exact ID.

Do not claim publication without a successful response.

MCP server referenced without pinned version: 'npx skills'.

Medium
Category
MCP Rug Pull
Confidence
90% confidence
Finding

The installation command runs the unpinned npm package skills through npx, allowing executable code to change between installations. A compromised future package release could execute with the installing user's privileges; this is a supply-chain exposure, not an MCP-server issue as the finding labels it. The repository reference is also not fixed to an immutable revision.

Content

No source excerpt is available for this finding.