ethskills

Fail

Audited by Gen Agent Trust Hub on May 9, 2026

Risk Level: CRITICALEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides instructions for the agent to execute shell commands for project scaffolding (npx create-eth@latest), package management (yarn install), and tool verification.
  • [EXTERNAL_DOWNLOADS]: Fetches installation scripts and configuration data from well-known and reputable technology organizations, including Noir-lang and AztecProtocol.
  • [REMOTE_CODE_EXECUTION]: Includes specific commands for downloading and executing setup scripts via curl | bash to install verified Ethereum development toolchains (nargo, bbup).
  • [DATA_EXFILTRATION]: Implements a feedback submission system that sends data to https://ethskills.com/api/feedback. This is governed by explicit safety instructions requiring the agent to strip PII and obtain human approval before transmission.
  • [PROMPT_INJECTION]: The skill relies on fetching and processing external markdown content from ethskills.com, which represents an indirect prompt injection surface. The impact is minimized by instructional guardrails and human-in-the-loop review patterns.
Recommendations
  • CRITICAL: 1 infected file(s) detected - DO NOT USE
  • Contains 1 malicious URL(s) - DO NOT USE
Audit Metadata
Risk Level
CRITICAL
Analyzed
May 9, 2026, 05:43 PM