weekly-market-brief-generator

Warn

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill requires the execution of shell commands to manage its environment and run analytical tools, specifically for activating virtual environments and running Python scripts as defined in references/data-queries.md.
  • [REMOTE_CODE_EXECUTION]: The skill executes Python scripts from a sibling directory (../findata-toolkit-cn/scripts/views_runner.py). Because this code is located outside the audited skill package, it constitutes the execution of unverified external code that can be modified independently of the skill.
  • [EXTERNAL_DOWNLOADS]: The environment preparation process involves running pip install -r ../findata-toolkit-cn/requirements.txt, which downloads and installs third-party packages from external registries like PyPI.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests and summarizes untrusted financial market data from external providers.
  • Ingestion points: Real-time market data (indices, industry boards, and fund flows) fetched via the views_runner.py tool as specified in references/data-queries.md.
  • Boundary markers: Absent. The skill does not implement delimiters or system instructions to ignore potentially malicious content embedded in the fetched financial data.
  • Capability inventory: The skill has permissions to execute shell commands and run Python scripts (documented in references/data-queries.md).
  • Sanitization: There is no evidence of data validation or sanitization logic to inspect external content before it is processed by the model.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 10, 2026, 03:23 AM