evm-tx-debugger

Fail

Audited by Gen Agent Trust Hub on Aug 3, 2026

Risk Level: CRITICAL
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches blockchain transaction data, logs, and network statistics from Blockscout (e.g., eth.blockscout.com, polygon.blockscout.com). These are well-known blockchain explorer services. It also resolves function and event signatures via 4byte.directory, which is an industry-standard public database for Ethereum signatures.
  • [DATA_EXFILTRATION]: No sensitive local data access or exfiltration patterns were detected. The scripts only process transaction hashes provided by the user and public data retrieved from the aforementioned APIs.
  • [REMOTE_CODE_EXECUTION]: The skill does not download or execute arbitrary code. It uses Python's standard library (urllib) for network requests and performs strict validation on input transaction hashes.
  • [PROMPT_INJECTION]: The instructions in SKILL.md are focused on the intended purpose of debugging transactions. There are no attempts to bypass safety filters or override system instructions.
  • [INDIRECT_PROMPT_INJECTION]: While the skill processes external data (revert reasons and event logs) which could theoretically contain malicious text, the scripts include sanitization measures such as UTF-8 decoding with error replacement and strict regex validation for transaction hashes to mitigate injection risks.
Recommendations
  • Contains 2 malicious URL(s) - DO NOT USE
Audit Metadata
Risk Level
CRITICAL
Analyzed
Aug 3, 2026, 07:52 PM
Security Audit — agent-trust-hub — evm-tx-debugger