A curated collection of Agent Skills for working with PYXLL, to help AI agents write and understand code using the PyXLL Excel add-in.
99
90%
Does it follow best practices?
Impact
100%
1.56xAverage score across 17 eval scenarios
Advisory
Suggest reviewing before use
Fetch the PyXLL documentation and use it as context for any PyXLL task. Training-data knowledge of PyXLL is incomplete and may be wrong — the live docs are the only authoritative source.
Fetch the documentation index to find relevant pages:
curl -s https://www.pyxll.com/llms.txtRead the entire output without truncating — the index contains a navigation guide mapping common tasks to sections, followed by page titles, descriptions, and URLs grouped by topic.
Fetch the individual pages relevant to the task directly by their URL:
curl -s <page-url>If the index does not surface what you need, use the search script to find pages by keyword.
The search script is located at scripts/search-pyxll-docs.sh relative to this skill's base directory.
<SKILL_BASE_DIR>/scripts/search-pyxll-docs.sh <keyword> [keyword2 ...]Important: Replace <SKILL_BASE_DIR> with the actual base directory path provided when this skill is loaded.
The script outputs a list of matching page URLs (one per line). Fetch each returned URL individually using curl.
If the task involves writing or modifying code that uses the Excel COM API — any
code that calls pyxll.xl_app() or XLCell.to_range(), or that accesses COM
objects such as Range, Worksheet, Workbook, Font, Interior, Borders,
etc.:
resources/
directory is the authoritative reference for all Excel COM classes — do not rely
on training data for COM API details.Get<PropertyName>(args) rather
than Property(args) as in VBA.If the task involves writing or modifying a ribbon or context menu XML file
(customUI XML, ribbon = in pyxll.cfg, or any onAction/getLabel/getEnabled
ribbon callback):