bottleneck-hunter
Pass
Audited by Gen Agent Trust Hub on Nov 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill establishes an attack surface by ingesting and acting upon untrusted data from external web sources.
- Ingestion points: The skill uses the WebSearch tool to gather industry news, capacity constraints, and supplier data which is then processed by the agent to form investment logic.
- Boundary markers: There are no explicit instructions or delimiters to isolate untrusted web content or to prevent the agent from obeying instructions potentially embedded in that content.
- Capability inventory: The skill allows the agent to execute shell commands via the Bash tool and write files to the local repository, which could be exploited if malicious instructions are ingested.
- Sanitization: No data validation or sanitization protocols are defined for the information retrieved from the internet.
- [COMMAND_EXECUTION]: The skill facilitates the execution of system commands and internal Python scripts to perform context verification and financial analysis.
- Evidence: The instructions direct the agent to run the 'date' command and execute local repository tools including 'python3 tools/financial_rigor.py' and 'python3 tools/twstock_data.py'.
Audit Metadata