inscription-queue-watcher

Pass

Audited by Gen Agent Trust Hub on Apr 30, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill performs read-only network operations to the author's official domain (aibtc.news) and an established Bitcoin infrastructure service (mempool.space).
  • [PROMPT_INJECTION]: The skill processes data from external APIs, creating a potential surface for indirect prompt injection. However, this is assessed as safe due to following factors:
  • Ingestion points: Brief information and transaction statuses are fetched via HTTP from aibtc.news and mempool.space (implemented in lib.ts).
  • Boundary markers: The skill produces structured JSON reports with defined schemas, which acts as a clear boundary for the agent when interpreting results.
  • Capability inventory: The skill is strictly observational and does not contain logic for file system writes, arbitrary command execution, or autonomous remediation.
  • Sanitization: The skill performs validation on inscription IDs and BTC addresses using specific regular expressions (parseInscriptionId and BECH32_BTC_ADDRESS).
  • [SAFE]: No obfuscation, hardcoded credentials, or dangerous command execution patterns were detected in the source code or instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 30, 2026, 11:12 PM