ib-deck

Pass

Audited by Gen Agent Trust Hub on Aug 2, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill correctly identifies and utilizes the vendor's own Daloopa MCP and associated tools for data retrieval. All external links point to the author's verified domain (daloopa.com).
  • [SAFE]: The skill incorporates specific constraints to prevent the hallucination of firm names, ensuring professional standards are maintained.
  • [SAFE]: The skill references a helper script named chart_generator.py for visualizing financial data, which serves as a local utility for the generation process.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data which could potentially contain malicious instructions.
  • Ingestion points: SEC filings and web search results are ingested during the Phase 2 research process.
  • Boundary markers: The skill lacks explicit delimiters or instructions for the model to ignore potential commands embedded within external data.
  • Capability inventory: The agent has the ability to generate complex HTML presentations and use discovery tools.
  • Sanitization: No evidence of input validation or sanitization for external content is provided.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 2, 2026, 07:47 PM
Security Audit — agent-trust-hub — ib-deck