scienceworld-tool-user

Uses a tool from inventory on a target object or location to perform a specific environmental interaction, such as digging, cutting, or measuring. Use this skill when a task requires modifying the environment or manipulating materials with a tool (e.g., using a shovel to dig soil, a thermometer to measure temperature, or an axe to cut wood). Takes the tool and target as inputs and outputs the result of the interaction.

Quality

93%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

87%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is concise, actionable, and well-structured with a clear worked example, and the bundle includes a real one-level reference. The only gap is the absence of an explicit validation/feedback loop for a potentially destructive environmental action.

Suggestions

Add an explicit verification step after 'Interpret Result' (e.g., if the observation indicates failure or an unexpected state, re-confirm the target name or re-acquire the tool and retry) to create a validate-fix-retry feedback loop.

Link the existing references/example_trajectory.md from the body (e.g., a 'See also' line) so the one-level reference is clearly signaled rather than present-but-unlinked.

Dimension	Reasoning	Score
Conciseness	Lean and efficient with tight sections, a compact actions table, and no over-explanation of concepts Claude already knows; every token earns its place.	3 / 3
Actionability	Provides fully executable commands (`pick up OBJ`, `use thermometer on metal fork`) plus a worked example with the actual observation string, making guidance copy-paste ready.	3 / 3
Workflow Clarity	The 4-step sequence is clearly ordered with an inventory prerequisite, but for an environment-modifying (potentially destructive) action there is no explicit validate/retry feedback loop after "Interpret Result", capping it at 2 per the rubric guideline.	2 / 3
Progressive Disclosure	A short single-purpose skill under 50 lines with well-organized sections and a real one-level-deep reference file (references/example_trajectory.md) present in the bundle, meeting the simple-skill bar for full marks.	3 / 3
	Total	11 / 12 Passed

Description

100%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is specific, trigger-rich, and explicitly covers both what the skill does and when to use it, with concrete tool-target examples that distinguish it from neighboring skills. It is concise without padding.

Dimension	Reasoning	Score
Specificity	Lists multiple specific concrete actions ("digging, cutting, or measuring") with concrete tool-target pairs ("shovel to dig soil", "thermometer to measure temperature", "axe to cut wood"), matching the multi-action anchor rather than a vague domain-only statement.	3 / 3
Completeness	Explicitly answers both what ("Uses a tool from inventory on a target object or location") and when ("Use this skill when a task requires modifying the environment or manipulating materials with a tool"), satisfying the explicit-trigger bar.	3 / 3
Trigger Term Quality	Covers natural terms a user would say in this domain ("digging", "cutting", "measuring", "modifying the environment", named tools), giving good trigger coverage rather than generic jargon.	3 / 3
Distinctiveness Conflict Risk	Anchored to the distinct `use TOOL on TARGET` niche with concrete tool examples, making it unlikely to trigger for unrelated skills despite the broadly-worded "modifying the environment" phrasing.	3 / 3
	Total	12 / 12 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: zjunlp/SkillNet
Path: experiments/src/skills/scienceworld/scienceworld-tool-user/SKILL.md
Commit: 7bb9a48

Reviewed: 1 day ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.