etherscan-contract-review

Pass

Audited by Gen Agent Trust Hub on Aug 18, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [UNVERIFIABLE_DEPENDENCIES_AND_REMOTE_CODE_EXECUTION]: The skill guides the user to install the etherscan-cli tool from the official vendor repository (github.com/etherscan/etherscan-cli). This is an author-provided resource and the skill requires explicit user consent and verification before any installation occurs.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted smart contract source code and metadata. It implements robust mitigations by instructing the agent to treat all retrieved artifacts as data rather than instructions and provides strict path normalization rules to prevent directory traversal during bundle reconstruction.
  • [DATA_EXPOSURE_AND_EXFILTRATION]: While noting that the etherscan-cli tool stores API keys locally, the skill proactively instructs the user never to share keys in chat and mandates that the agent does not inspect or echo the saved credentials.
  • [COMMAND_EXECUTION]: Uses CLI commands for read-only contract retrieval. The workflow explicitly forbids state-changing actions, transaction broadcasting, and wallet signing.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 18, 2026, 06:24 AM
Security Audit — agent-trust-hub — etherscan-contract-review