The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill requires cloning a project repository from github.com/Timeverse/My-TW-Coverage and installing several third-party libraries (yfinance, pandas, tabulate) via pip.
[COMMAND_EXECUTION]: The skill provides examples and triggers that instruct the agent to use subprocess.run or direct shell execution to run local Python scripts (e.g., add_ticker.py, update_financials.py, discover.py) to manage the database.
[INDIRECT_PROMPT_INJECTION]: The skill builds a knowledge graph by ingesting financial reports and external research data, which presents a surface for indirect prompt injection.
Ingestion points: Research reports stored in Pilot_Reports/ and JSON data provided via the update_enrichment.py script.
Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore potentially malicious instructions embedded within the research reports.
Capability inventory: The skill possesses the capability to execute shell commands via subprocess.run and perform network operations using the yfinance library.
Sanitization: While the skill includes an audit_batch.py script for quality control, it focuses on content completeness and formatting rather than security sanitization or filtering of prompt injection patterns.

taiwan-equity-research-coverage