investor-research
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted data gathered from web searches and user-provided JSON configurations, which is then interpolated into a markdown report template without sanitization.\n
- Ingestion points: Startup profile information and investor data gathered via web searches (as instructed in
SKILL.md) and the JSON input file processed byscripts/research_investors.py.\n - Boundary markers: The Python script lacks boundary markers or explicit instructions to ignore potentially malicious embedded content within the JSON values during interpolation.\n
- Capability inventory: The skill utilizes a Python script to write files to the local system and requires the agent to perform web searches and provide inputs to that script.\n
- Sanitization: No sanitization, escaping, or validation of the data is performed before it is written into the final markdown report.\n- [COMMAND_EXECUTION]: The skill requires the agent to execute a local Python script (
scripts/research_investors.py) to format data and generate the report. This involves local command execution, which is the primary intended functionality of the skill.
Audit Metadata