Skip to content

The commands of this book

Every command takes --json. A person in the last column means: the command refuses a process of an agent and the name of an agent.

Command What it does Who Chapter
sapling new <study> --protocol usdm.json --sdtm <folder> the study folder and the first draft of the SAP anyone 4
sapling check the five passes of the compiler anyone 4
sapling eval · sapling round --note "..." · sapling eval history the scorecard; a round is saved and compared with the last one anyone 5, 10
sapling next · sapling status · sapling help the next thing to do; your one screen; your commands anyone 5
sapling agent · sapling agent --print your AI agent in the study folder, with its request you 6
sapling show <block> · sapling blocks · sapling find <words> a block with its protocol sentence; the blocks; what the data have and where it is used anyone 7, 13
sapling block check <file> --after <block> [--write] one block, checked in its place, written only when it compiles the agent (or you) 7
sapling generate sensitivity (and other kinds) a conventional block in its section the agent (or you) 8
sapling judge the dossier, out/eval/judge.html anyone 9
sapling revise <block> "..." --by NAME a note that the agent answers first a person 9
sapling reply <block> "..." the agent’s answer to your note the agent 9
sapling agree <block> --by NAME · --kind K agreement to the text of a block, pinned to a hash a person 9, 20
sapling decide <n> "..." --by NAME · --file answers.tsv the answer to an open decision a person 10
sapling approve --by NAME --role ROLE the approval, only when the plan is READY FOR APPROVAL a person 20
Command What it does Who Chapter
sapling build the structured SAP, its PDF, the specifications, the shells, the programs anyone 11
sapling run the ADaM data and the tables, in R through DataStep Studio for R anyone 12
sapling adam conform the ADaM data against 27 rules of the ADaMIG v1.2 anyone 13
sapling r programs · sapling r run · sapling r check the readable R programs, their run, every number against the engine’s anyone 16
sapling diagnose [OUTPUT|DATASET] what the last run did, each failure with its trace anyone 17
sapling fix · sapling accept <id> --by NAME one tested correction; a person accepts it the agent; a person accepts 17
sapling blind --seed S the real data with a dummy treatment anyone 18
sapling unblind --restore --by NAME · --randomization <list> --by NAME the real treatment back a person 18
sapling dashboard · sapling dashboard --check the study dashboard; its gate anyone 19
sapling final · sapling final --make the state of the four products; make them and judge them anyone 20
sapling new <study> --indication <id> --seed S a made-up study of an indication, with dummy data anyone 21
sapling doctor what the computer needs, with the fix anyone 2

The full reference of every command is in the Sapling manual.

Found an error or have a question? Write to datastep.studio@gmail.com· The plan is written in Sapling; every R program runs in DataStep Studio for R.

Data: the CDISC pilot study (cdiscpilot01). The protocol as CDISC USDM from the CDISC 360i project (MIT licence); the SDTM data from the PhUSE scripts repository (MIT licence). Public example data: no real patient.

R is a project of the R Foundation. SAS is a registered trademark of SAS Institute Inc. CDISC, SDTM, ADaM and USDM are trademarks of the Clinical Data Interchange Standards Consortium. Claude is a trademark of Anthropic. This book is not affiliated with or endorsed by any of them.

© 2026 DataStep Labs. All rights reserved.