The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local scripts using shell commands to perform its primary function. Specifically, it runs node "$HOME/.claude/skills/stock-recap/scripts/gather.mjs" to collect data and .venv/bin/python scripts/render_chart.py to generate visual candlestick charts.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface (Category 8). It ingests untrusted data from multiple external sources, including TraderDaddy Pro APIs, TickerTrace, and CBOE listings. This data is compiled into a report.md file, which the agent is then explicitly instructed to read and interpret.
Ingestion points: The agent reads data from report.md, which is populated with content from external APIs (/api/agent/bounce-finder, /api/earnings-gap, TickerTrace 13F data).
Boundary markers: There are no explicit delimiters or instructions provided to the agent to ignore potentially malicious instructions embedded within the processed financial data.
Capability inventory: The skill has the capability to execute shell commands (node, python) and read files from the local filesystem.
Sanitization: There is no evidence of sanitization or filtering of the external API content before it is presented to the agent in the report.
[EXTERNAL_DOWNLOADS]: The documentation provides instructions to install standard Python libraries (mplfinance, pandas) from public registries to support the chart rendering functionality.

stock-recap