scienceworld-object-locator

Searches for a specific target object across multiple rooms by systematically teleporting to likely locations and examining each room. Use this skill when you need to find an object whose exact location is unknown. Iterates through candidate rooms using teleport and look around, checks for the object, and continues until found, returning the object's location.

Quality

69%

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

Fix and improve this skill with Tessl

tessl review fix ./experiments/src/skills/scienceworld/scienceworld-object-locator/SKILL.md

Quality

Content

71%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 clear, actionable, well-sequenced skill body that mostly respects Claude's competence. Its main weakness is progressive disclosure: the provided reference bundle is never signaled from the overview, and a referenced script is missing.

Suggestions

Add a short 'References' section linking to the existing bundle files (e.g., 'For full action details, see references/action_guide.md; for a worked example, see references/example_trajectory_breakdown.md').

Either bundle the `prioritize_rooms.py` script under ./scripts/ or replace that step with an inline default room-priority list so step 2 is actually executable.

Fold the `examine`-to-confirm guidance into the search loop as an explicit verification checkpoint so the workflow's validation step is visible in the Core Process.

Dimension	Reasoning	Score
Conciseness	The body is lean and sectioned with no padding about what ScienceWorld or teleport means; the only slack is minor restatement between 'Core Process' and 'Notes & Best Practices' (e.g., 'look around immediately after teleporting').	4 / 5
Actionability	Concrete, copy-paste-ready commands are given (`teleport to <ROOM_NAME>`, `look around`, `examine <OBJECT>`) with a worked example, but step 2 directs use of a `prioritize_rooms.py` script that is not present in any ./scripts/ bundle, leaving a real execution gap.	4 / 5
Workflow Clarity	The Core Process is a clearly numbered sequence with an explicit stop condition and a not-found failure path; the only gap is that identity verification (`examine` for ambiguous objects) lives in Notes rather than being wired in as a workflow checkpoint.	4 / 5
Progressive Disclosure	The body is well-sectioned, but it never links to the three actual bundle files in ./references/ (action_guide.md, action_primer.md, example_trajectory_breakdown.md) whose detailed content clearly belongs one level deep, and it instead references a non-existent prioritize_rooms.py script.	3 / 5
	Total	15 / 20 Passed

Description

67%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 solid description that states both capability and an explicit use-trigger with concrete actions. It is held back by limited trigger-term synonyms and the absence of the ScienceWorld domain name, which would sharpen distinctiveness.

Suggestions

Broaden trigger terms to include natural synonyms users say, e.g. 'locate', 'search for', or 'find where the <object> is'.

Name the ScienceWorld environment in the description so the skill is clearly distinguishable from generic search/navigation skills.

Tighten the repeated teleport/examine phrasing into a single crisp action list to lift specificity toward comprehensive coverage.

Dimension	Reasoning	Score
Specificity	Names several concrete actions — 'teleporting to likely locations', 'examining each room', 'teleport and look around', 'checks for the object', 'returning the object's location' — giving good coverage of the locate task, though teleport/look-around are restated rather than expanded with new actions.	4 / 5
Completeness	Clearly answers 'what' (search rooms via teleport/look around, iterate, return location) and 'when' ('Use this skill when you need to find an object whose exact location is unknown'), but the trigger could be more specific about the ScienceWorld context.	4 / 5
Trigger Term Quality	Offers one natural trigger phrasing ('find an object whose exact location is unknown') but misses common synonyms users would say such as 'locate', 'search for', or 'where is the X', leaving keyword coverage thin.	3 / 5
Distinctiveness Conflict Risk	The teleport-and-examine-rooms niche is fairly distinct, but because the description never names the ScienceWorld environment, there is minor overlap risk with generic navigation/search skills.	4 / 5
	Total	15 / 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-object-locator/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.