The Agent Skills Directory

[SAFE]: The skill performs legitimate financial analysis operations using data from well-known and reputable services. All network activity is directed toward established platforms like Yahoo Finance and Google News, consistent with the skill's stated purpose.
[PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it ingests untrusted data from external sources, specifically news headlines and company metadata.
Ingestion points: External text data enters the agent context via the scripts/analyze_stock.py script through RSS feeds and financial APIs.
Boundary markers: The script output is presented to the agent without specific delimiters or instructions to ignore instructions that might be embedded in the fetched text.
Capability inventory: The skill's capabilities are restricted to fetching data and managing a local JSON-based portfolio database; it has no access to sensitive system resources or arbitrary code execution tools.
Sanitization: Content fetched from news feeds and ticker metadata is not sanitized or escaped before presentation to the agent.
[COMMAND_EXECUTION]: The skill uses uv run to execute its internal analysis and portfolio management scripts. These scripts are self-contained and do not accept arbitrary user input for shell execution.
[EXTERNAL_DOWNLOADS]: The skill fetches data and configuration from well-known services including Yahoo Finance, Google News, and CNN's Fear & Greed index. It utilizes standard, open-source Python libraries for data processing and API interaction.

stock-analysis