papyrus-optimization
Papyrus Optimization
Overview
This skill is a read-and-assess investigation flow for Papyrus (.psc) source. It lets you read a
script and classify each part as π΄ broken, π‘ suboptimal, or π’ clean, name what makes
a part heavy, and give the specific fix β so you catch performance problems a signature lookup can't
see. It is the cost-and-habits complement to the papyrus-reference skill (which gives function
signatures): that one tells you what a call is; this one tells you what it costs and whether to
make it. The deep cost model, exact numbers, full technique catalogue, and citations live in
references/papyrus-performance-reference.md β load it
when you need the why behind a verdict; the workflow below is enough for most reviews.
It reads script text rather than houseCARL's record/conflict surfaces, so it composes with normal
file reading (or housecarl_bsa_extract + the decompiler when the script is packed) β no L2 query
surface is required.