The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes untrusted external data from financial websites, earnings reports, and analyst call transcripts. This creates a surface for indirect prompt injection where malicious instructions embedded in these documents could influence the agent's behavior.
Ingestion points: External financial data sources including Seeking Alpha, IR pages, and news sites.
Boundary markers: No specific delimiters or "ignore" instructions are used when processing the external data.
Capability inventory: The skill executes shell commands (python3 tools/financial_rigor.py, python3 tools/report_audit.py) and writes reports to the local file system.
Sanitization: No explicit sanitization or validation of the content extracted from external documents is specified before it is used as arguments in shell commands or included in reports.
[COMMAND_EXECUTION]: The skill uses local shell commands to execute Python scripts for financial data verification and auditing. These scripts are part of the repository's toolset.
Evidence: Steps 2.3 and 8 use python3 tools/financial_rigor.py and python3 tools/report_audit.py to validate financial metrics and audit generated reports.
[EXTERNAL_DOWNLOADS]: The skill fetches financial data from several well-known technology and financial service providers.
Evidence: Step 1 describes fetching original materials from the SEC EDGAR (sec.gov), HKEX (hkexnews.hk), Seeking Alpha, and Yahoo Finance. These are established services and their use for financial analysis is standard.

earnings-review