crypto-onchain-data

Pass

Audited by Gen Agent Trust Hub on Aug 20, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local Python script to perform its primary data retrieval tasks.
  • Evidence: Command python3 .agents/skills/crypto-onchain-data/onchain_fetch.py is invoked to gather MVRV-Z, NUPL, and other metrics.
  • [EXTERNAL_DOWNLOADS]: The skill performs network requests to several external domains to fetch financial and on-chain data.
  • Evidence: Requests are made to bitcoin-data.com, farside.co.uk, sosovalue.com, coinglass.com, and newhedge.io.
  • [DATA_EXPOSURE]: The skill accesses the file system to write a cache file in a directory path outside its own skill folder.
  • Evidence: The script onchain_fetch.py navigates to ../../../crypto/news/onchain_cache.json for persistent storage.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection due to the ingestion of untrusted data from multiple external crypto-market websites.
  • Ingestion points: Data is pulled from external URLs (bitcoin-data.com, farside.co.uk, etc.) and a local cache file.
  • Boundary markers: Absent. The instructions do not provide delimiters or warnings for the agent to ignore instructions that might be embedded in the fetched content.
  • Capability inventory: The skill can execute local Python scripts, perform network operations, and access the filesystem.
  • Sanitization: Absent. There is no evidence of filtering or validation of the content retrieved from external sources before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 20, 2026, 02:21 PM
Security Audit — agent-trust-hub — crypto-onchain-data