suitability-report-generator

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides explicit instructions in references/data-queries.md to execute shell commands for activating a Python virtual environment (source ../../.venv/bin/activate) and running a data runner script (python ../findata-toolkit-cn/scripts/views_runner.py). These actions are transparently documented as part of the data augmentation workflow.
  • [EXTERNAL_DOWNLOADS]: The environment setup process includes an instruction to install Python dependencies from a local requirements file (python -m pip install -r ../findata-toolkit-cn/requirements.txt). While this initiates external downloads through the package manager, the behavior is expected for a data-intensive financial analysis skill.
  • [PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection due to its core function of ingesting and summarizing untrusted external data into a compliance report.
  • Ingestion points: SKILL.md (Investment recommendations and client profile data provided by users).
  • Boundary markers: The instructions lack explicit boundary markers or delimiters to isolate user-provided text from the agent's internal logic.
  • Capability inventory: The skill utilizes shell and Python script execution capabilities (references/data-queries.md) to fetch market data.
  • Sanitization: No evidence of input validation or sanitization is present for the user-provided data inputs.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 03:23 AM