injective-evm-developer

Pass

Audited by Gen Agent Trust Hub on Apr 12, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Recommends the installation of the whackur/solidity-agent-toolkit skill, which is a third-party resource not maintained by the official vendor.\n- [EXTERNAL_DOWNLOADS]: Fetches Solidity contract templates and documentation from the official InjectiveLabs GitHub repository and documentation domains. These are trusted vendor resources.\n- [COMMAND_EXECUTION]: Provides scripts and command-line instructions for compiling, deploying, and verifying smart contracts using the Hardhat framework. These commands are executed by the user or agent in the project environment.\n- [PROMPT_INJECTION]: As a developer-focused skill that analyzes project files (.sol, .js, .ts), it is susceptible to indirect prompt injection if the project being analyzed contains malicious instructions.\n
  • Ingestion points: Project workspace files matching the extensions *.sol, *.ts, *.js, and config files like hardhat.config.* and foundry.toml.\n
  • Boundary markers: No specific delimiters or instructions to ignore embedded commands are provided in the guide.\n
  • Capability inventory: The skill enables the agent to execute contract deployment, verification, and interaction commands via standard developer tools.\n
  • Sanitization: Relies on the host environment and user verification for safety; no explicit sanitization logic is included in the skill instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 12, 2026, 10:16 AM