tres-tx-story

Pass

Audited by Gen Agent Trust Hub on Aug 1, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • External Data Ingestion (Indirect Prompt Injection): The skill processes data from blockchain records (such as contract names and wallet aliases) fetched from the TRES Finance API. As this information originates from public blockchain activity, it is technically user-influenced and could contain unexpected strings. These values are used to generate visual diagrams and dynamic prompts.
  • Ingestion Points: Data is ingested through the GraphQL transaction query in SKILL.md, specifically fields like displayName and contractName.
  • Boundary Markers: There are no explicit delimiters or instructions to ignore embedded commands when these values are used in subsequent steps.
  • Capability Inventory: The skill uses show_widget to render SVG content and sendPrompt to create interactive buttons for the user.
  • Sanitization: The instructions do not specify sanitization or escaping of the retrieved blockchain strings before they are interpolated into the UI or secondary prompts.
  • External Service Integration: The skill connects to a dedicated blockchain analysis service (https://ai.tres.finance/mcp). This connection is documented and essential for the skill's primary function of providing real-time transaction data.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 1, 2026, 04:44 AM
Security Audit — agent-trust-hub — tres-tx-story