Agents
Give the maintenance request a clear structure
BOTIE
Examples, experiments and workspace controls
Connect a tool's inputs to see what it produces. You can try a branch before connecting it to Result.
Try your own inputs
Use a copy of the example to try your own data. Keep the problem's assumptions in mind, and predict the effect of a change before checking the output. You can also add another branch to compare two ways of doing the calculation.
Context for this challenge
Label the question and manual as separate JSON fields and join the two text blocks with a newline. Keep their contents unchanged.
Botie needs to distinguish the application’s job from a technician’s question and an untrusted manual excerpt.
Inspect the component calculation
Join two text blocks(
First text: Label a value as JSON(
Data value: Technician question
Field name: question
)
Second text: Label a value as JSON(
Data value: Retrieved manual text
Field name: manual
)
)Examples and results
- Technician question
What should P-1 check?
- Retrieved manual text
M1: inspect the seal.
- Required result
{"question":"What should P-1 check?"} {"manual":"M1: inspect the seal."}
- Technician question
Is P-9 "safe"?
- Retrieved manual text
M9 has no inspection interval.
- Required result
{"question":"Is P-9 \"safe\"?"} {"manual":"M9 has no inspection interval."}
Try a fresh construction
This clears this draft, including its undo history. Earlier successful attempts stay recorded. Build again without opening hints to record an independent attempt.
Stuck? Inspect a worked construction
The tools below form one possible solution. You can assemble it, follow its signals, and undo that change to return to your own construction.
Assumptions and test precision
This checks request construction, not model obedience. JSON labels do not prevent prompt injection; later server-side tool checks enforce permissions. Numerical tests allow a difference of 0.001. Lists and structures must also have the right shape.
Your last tested pieces can start the full construction. An existing draft there keeps its own pieces.
The service ticket contains a line break
This check uses the pieces from “Give instructions and evidence separate messages”.
BOTIE
- Application-owned task
Explain the supplied source.
- Required response format
Include its ID.
- Technician question
Which part should P-3 check?
- Retrieved manual text
M3: inspect the "seal". Record the finding.
Use a number, text, or an array matching the requested result. For a matrix, put each row in brackets. Saving your prediction runs your current construction on these inputs and checks the earlier examples too.
Build a construction in the workspace above first. This case will test those pieces.
Which checks would still be needed if a model followed the malicious manual anyway?
Keep action permissions outside the prompt. Next, check the shape of the model’s proposed answer.
This checks request construction, not model obedience. JSON labels do not prevent prompt injection; later server-side tool checks enforce permissions.
Work through the Python challenges →