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

89%

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

90%Weight 40%Scale 1-5

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

A concise, highly actionable skill body with concrete commands and a worked example. Workflow sequencing is clear but light on validation/retry guidance, and the existing reference file is not surfaced from the body.

Suggestions

Add an explicit error-recovery step in the Core Workflow (e.g., if `use TOOL on TARGET` fails or returns no change, re-run `look around` to verify the target name and retry).

Link the example trajectory from the body (e.g., under the Example section: 'See [example_trajectory.md](references/example_trajectory.md) for a full end-to-end trace') so the reference is clearly signaled.

Note prerequisite validation explicitly (confirm tool is in inventory via `look around` / inventory check before executing `use`) to strengthen the workflow checkpoint.

Dimension	Reasoning	Score
Conciseness	Lean ~35-line body that assumes Claude's competence, avoids explaining what tools are, and lets every line earn its place.	5 / 5
Actionability	Provides concrete executable commands (`pick up OBJ`, `use TOOL on TARGET`, `look around`) plus a complete worked example with exact inputs and the observation output, covering common cases.	5 / 5
Workflow Clarity	A clear 4-step sequence (ensure tool, identify target, execute, interpret) with a light confirm step, but lacks explicit error-recovery or retry checkpoints for failed interactions.	4 / 5
Progressive Disclosure	Well-organized sections for a simple single-purpose skill, with a one-level-deep reference (references/example_trajectory.md), but that reference is not linked or signaled from the body.	4 / 5
	Total	18 / 20 Passed

Description

88%Weight 40%Scale 1-5

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

A strong, specific description that answers both 'what' and 'when' with concrete examples and an explicit use-when clause. Trigger term coverage and distinctiveness are good but could be sharper with more synonyms and tighter scoping.

Dimension	Reasoning	Score
Specificity	Names the domain and multiple concrete actions (digging, cutting, measuring) with specific examples (shovel to dig soil, thermometer to measure temperature, axe to cut wood), giving comprehensive coverage.	5 / 5
Completeness	Clearly states what it does ('Uses a tool... to perform a specific environmental interaction', takes tool and target as inputs, outputs result) and explicitly when to use it ('Use this skill when a task requires modifying the environment...').	5 / 5
Trigger Term Quality	Includes natural terms like 'modifying the environment', 'manipulating materials with a tool', and 'digging, cutting, or measuring', but lacks synonym variation and broader trigger phrasings a user might say.	4 / 5
Distinctiveness Conflict Risk	The tool-on-target environmental-interaction niche is mostly distinct, but 'use a tool to modify the environment' is broad enough to risk minor overlap with other inventory/action skills.	4 / 5
	Total	18 / 20 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: 076c6df

Reviewed: 6 days 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.