Scienceworld Measurement TakerSAFE
Create, Evaluate, and Connect AI Skills
Overview
Create, Evaluate, and Connect AI Skills
4501e739125cOBSERVED · 2026-10-12What it tells the agent
The instruction file, verbatim from the audited commit — this is the text the model reads, and the surface the audit's instruction layer examines. Quoted here so you can judge it without cloning anything.
--- name: scienceworld-measurement-taker description: Use when the agent needs to measure a quantitative property (temperature, weight, pH) of a target object or substance using a measurement tool. This skill covers acquiring the tool, preparing both tool and target with focus actions, executing the measurement via "use [TOOL] on [TARGET]", and interpreting the resulting value for conditional decisions such as classification or placement. --- # Skill: scienceworld-measurement-taker ## Purpose Measure a quantitative property of a target object or substance in the ScienceWorld environment using the appropriate measurement tool, then interpret the result for subsequent decision-making. ## When to Use - When a task requires obtaining a numerical reading (temperature, weight, pH) from an object or substance - When a measured value determines a conditional next action (e.g., classify based on temperature threshold) - When the agent needs to compare properties across multiple objects ## Workflow 1. **Identify and acquire the tool** -- Locate the correct measurement instrument (e.g., thermometer, scale) and `pick up` the tool. 2. **Focus on the tool** -- Execute `focus on [TOOL] in inventory` to confirm readiness. 3. **Identify and acquire the target** -- Locate the target object or substance and `pick up` the target. 4. **Focus on the target** -- Execute `focus on [TARGET] in inventory` to confirm readiness. 5. **Position for follow-up** -- If the task requires a follow-up action (e.g., placing in a bin), `teleport to` the appropriate location before measuring. 6. **Execute measurement** -- Use the tool on the target: `use [TOOL] on [TARGET]`. 7. **Interpret and act** -- Read the numerical result and execute the appropriate conditional action. ## Examples ### Example 1: Measure temperature and classify ``` > teleport to kitchen You teleport to the kitchen. > pick up thermometer You pick up the thermometer. > focus on thermometer in inventory You focus on the thermometer. > pick up unknown substance B You pick up the unknown substance B. > focus on unknown substance B in inventory You focus on the unknown substance B. > teleport to hallway You teleport to the hallway. > use thermometer on unknown substance B The thermometer measures a temperature of 56 degrees celsius. > move unknown substance B to orange box You move the unknown substance B to the orange box. ``` ### Example 2: Measure and compare two objects ``` > use thermometer on metal fork The thermometer measures a temperature of 23 degrees celsius. > use thermometer on glass cup The thermometer measures a temperature of 87 degrees celsius. ``` The glass cup (87) is hotter than the metal fork (23). ## Key Principles - **Tool first** -- Secure the measurement tool before handling the target. - **Focus before use** -- Always `focus on` both tool and target in inventory before measuring. - **Plan ahead** -- Identify where follow-up actions occur and position accordingly before measuring. - **Map thresholds** -- Clarify decision thresholds (e.g., "above 100.0" vs. "below 100.0") before executing the measurement.
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | NA |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- none-observed
- Network
- none-observed
- Shell
- none-observed
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (0)
No findings outside the package's declared scope.
Gates applied: no_behavioural_pass.
4501e739125cfull audit observations/trust-audit/skill/zjunlp__scienceworld-measurement-taker.json · Report an issue / request a re-scanAudit history
Every audit this skill has had.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-12 | 4501e739125c | SAFE | B | 89 | first audit |
Questions
What does the Scienceworld Measurement Taker skill do?
Create, Evaluate, and Connect AI Skills
Is Scienceworld Measurement Taker safe to install?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean skill reads B.
What can Scienceworld Measurement Taker access on my machine?
The audit observed no filesystem, network or shell use at all in its source.
How current is this page?
The grade is for one exact copy of the source (4501e739125c), read on 2026-10-12. The repository is watched, and a new audit runs when it changes — this is the first audit.