The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill implements automated update checks and retrieves financial data from external sources.
scripts/common.py fetches update information from the official repository at github.com/pseudo-longinus/quant-buddy-view.
scripts/materialize_new_asset_csv.py downloads financial CSV data from well-known cloud storage providers.
[REMOTE_CODE_EXECUTION]: A self-update mechanism is present that downloads and installs code.
scripts/self_update.py programmatically replaces local skill files with code downloaded from the vendor's GitHub repository. This is an intended feature for maintenance from a trusted source.
[COMMAND_EXECUTION]: The skill makes extensive use of sub-processes to coordinate quantitative analysis and validation.
scripts/qbs_bridge.py executes sibling skills like quant-buddy-skill via the command line.
scripts/verify_page.mjs and scripts/capture_rendered_html.mjs spawn Node.js processes to utilize Playwright for browser rendering and layout validation tasks.
[INDIRECT_PROMPT_INJECTION]: The skill has a significant attack surface due to the ingestion and processing of untrusted user files.
scripts/prepare_existing_file.py and scripts/capture_file_snapshot.mjs process user-supplied HTML, PDF, and image files for inclusion in dashboards.
The logic incorporates these data into a pipeline that includes network and file system access.
[DYNAMIC_EXECUTION]: Dashboard generation involves the creation of executable content at runtime.
scripts/build_dashboard.py and scripts/compose_page.py assemble HTML and JavaScript code blocks that are executed in the user's browser to provide real-time data updates.

quant-buddy-view