Skills
Navigate
skills/fuzzystodd/the-nexus-protocol-token-dao/use-arc/Gen Agent Trust Hub

use-arc

Fail

Audited by Gen Agent Trust Hub on Oct 28, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill includes a command to install the Foundry development toolkit by downloading and executing a shell script from a remote URL (https://foundry.paradigm.xyz).
  • [INDIRECT_PROMPT_INJECTION]: The skill encourages the agent to prioritize reading external documentation and configuration from official sources, which presents a surface for instructions embedded in external content to influence the agent.
  • Ingestion points: Reference links to llms.txt files on docs.arc.network and developers.circle.com in SKILL.md.
  • Boundary markers: No specific delimiters or instructions are provided to the agent to treat external content as untrusted or to ignore embedded instructions.
  • Capability inventory: The skill enables shell command execution via the forge CLI tool for contract deployment and interaction.
  • Sanitization: The instructions do not implement content validation or sanitization for the data retrieved from external links.
  • [COMMAND_EXECUTION]: The skill provides patterns for executing shell commands to deploy smart contracts using the Forge utility.
Recommendations
  • HIGH: Downloads and executes remote code from: https://foundry.paradigm.xyz - DO NOT USE without thorough review
Audit Metadata
Risk Level
HIGH
Analyzed
Oct 28, 2026, 02:26 AM