earnings-reaction-analyzer
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructions direct the agent to perform environment setup and script execution using shell commands. This includes activating a virtual environment (
source ../../.venv/bin/activate) and running various Python scripts (e.g.,stock_data.py,sec_edgar.py) located in a sibling toolkit directory (../findata-toolkit/scripts/). These are standard operations for a data analysis toolkit. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external data from stock market providers and SEC filings, which represents a potential attack surface for indirect prompt injection.
- Ingestion points: Data is retrieved from external market sources and SEC EDGAR filings via scripts referenced in
references/data-queries.md. - Boundary markers: The instructions do not specify explicit delimiters or "ignore instructions" wrappers for the content retrieved from these external sources.
- Capability inventory: The skill uses command execution capabilities to run data retrieval and analysis scripts as documented in
references/data-queries.md. - Sanitization: There is no evidence of explicit sanitization or filtering of the external data within the skill's markdown instructions; it relies on the execution logic of the external Python scripts.
Audit Metadata