scienceworld-target-locator

This skill determines the most likely location for a target object based on domain knowledge and environmental clues. Use when the agent needs to find a specific item (like an animal) but it is not in the current room. It analyzes the environment description and suggests a room to teleport to for further investigation.

Quality

85%

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

92%

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, concrete, and well-sequenced with an explicit verify-and-retry checkpoint, scoring top marks on conciseness, actionability, and workflow clarity. The one gap is progressive disclosure: an existing reference file is neither linked nor signaled from the body, and its content is duplicated inline.

Suggestions

Link to the existing reference from the body, e.g. 'For the full type-to-room mapping, see [target_location_heuristics.md](references/target_location_heuristics.md)' instead of duplicating the table inline.

Either move the detailed heuristic table into the reference and keep a short summary in the body, or drop the reference file if the inline table is intended to be the canonical source, to avoid the duplicate-content/redundant-reference mismatch.

Dimension	Reasoning	Score
Conciseness	The body is lean and assumes competence: it never explains what ScienceWorld or teleport is, jumping straight into a procedure, a tight heuristic table, and one example, with no padding to tighten.	3 / 3
Actionability	Provides fully executable commands ('teleport to <ROOM>', 'look around') plus a concrete type-to-room mapping table and a worked example with exact outputs, reaching copy-paste-ready guidance.	3 / 3
Workflow Clarity	A clear four-step sequence includes an explicit verification checkpoint with a feedback loop (step 4: 'look around to verify... If not, try the next likely room'), matching the validation-and-recovery anchor.	3 / 3
Progressive Disclosure	A reference file (references/target_location_heuristics.md) exists but is never signaled or linked from the body, and its heuristic content is duplicated inline in the table; references are present but not clearly navigated.	2 / 3
	Total	11 / 12 Passed

Description

77%

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 clear, third-person description that states concrete actions and an explicit 'Use when' trigger, answering both what and when well. Trigger-term coverage is somewhat narrow and the niche could be more distinct from generic search/locate skills.

Suggestions

Broaden trigger terms to include natural variations users would say, e.g. 'where is X,' 'locate,' or 'I need to find,' rather than only 'find a specific item.'

Tighten distinctiveness by naming the ScienceWorld domain and the teleport mechanism in the description so it does not overlap with generic object-search skills.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions ('determines the most likely location,' 'analyzes the environment description,' 'suggests a room to teleport to'), matching the multiple-specific-actions anchor rather than the partial-actions level below.	3 / 3
Completeness	Explicitly states what the skill does and provides an explicit 'Use when the agent needs to find a specific item' trigger, satisfying both halves; it is not the implied-only level below.	3 / 3
Trigger Term Quality	Includes a 'Use when...' clause with natural phrasing ('find a specific item (like an animal)'), but coverage of trigger variations is narrow and misses common forms like 'where is' or 'locate'; not the broad coverage of the top anchor.	2 / 3
Distinctiveness Conflict Risk	The teleport/room-not-present niche is fairly specific, but generic terms ('target object,' 'find a specific item') could overlap with general search skills, so it is not maximally distinct.	2 / 3
	Total	10 / 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-target-locator/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.