CtrlK
BlogDocsLog inGet started
Tessl Logo

alfworld-tool-locator

Searches for a specified tool or device (e.g., a desklamp, knife, or sponge) within the ALFWorld environment by checking relevant surfaces. Use when you need a tool to interact with another object as part of a task but the tool is not in your inventory or immediate vicinity. Takes a tool name as implicit input and navigates to likely storage spots (sidetables, shelves, countertops) until the tool is found.

71

Quality

89%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

85%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 tight, highly actionable skill body with an explicit branching workflow and feedback loops for error recovery. Its one real structural weakness is progressive disclosure: two reference files exist in the bundle, but the body never signals them and instead inlines their content, leaving the references undiscoverable.

Suggestions

Add clearly signaled, one-level-deep links to the existing bundle files (e.g., under Workflow: 'Full tool-category priorities: see [tool_categories.md](references/tool_categories.md)' and under Action Format: 'All available actions: see [alfworld_actions.md](references/alfworld_actions.md)'), and trim the inlined duplicated lists accordingly.

Trim the example's first two Thought lines that merely restate the workflow's priority rules, keeping the transcript focused on the Action/Observation alternation.

Clarify the 'verify the receptacle name matches the environment scan' recovery step with the concrete fix (e.g., re-issue 'go to' with the exact receptacle name and index from the scan).

DimensionReasoningScore

Conciseness

The body is lean (~40 lines) with no explanation of concepts Claude already knows — every section (Workflow, Action Format, Error Recovery, Example) carries operational content. It falls short of a 5 because the example's opening thoughts ('I need to find a desklamp. Desklamps are usually on sidetables or desks.') restate what the workflow already says and could be trimmed.

4 / 5

Actionability

Guidance is fully executable in ALFWorld's exact command format: 'go to {receptacle}', 'take {tool} from {receptacle}', 'toggle {tool}', plus a copy-paste-ready Thought/Action/Observation transcript covering the common case. The concrete receptacle-type priority lists per tool category make the guidance immediately actionable.

5 / 5

Workflow Clarity

The workflow is a clearly sequenced loop with explicit checkpoints: prioritize by tool type → 'go to {receptacle}' → 'Inspect the observation for the target tool name' → stop if found, advance if not, expand if exhausted. Error Recovery provides feedback loops ('If "Nothing happened": verify the receptacle name'; expand search when exhausted). This is not a destructive/batch operation, so no validation cap applies, and the recovery loops satisfy the 5 anchor.

5 / 5

Progressive Disclosure

The body is well-sectioned internally, but the bundle contains references/alfworld_actions.md and references/tool_categories.md and neither is ever mentioned or linked from the body — the tool-location priority lists in the Workflow section duplicate references/tool_categories.md inline instead of pointing to it, and the action formats duplicate references/alfworld_actions.md. This matches the 3 anchor ('references present but not clearly signaled; content that should be separate is inline'), not the simple-skill 5 exception, because external reference files exist and are orphaned.

3 / 5

Total

17

/

20

Passed

Description

87%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 description: concrete third-person actions, explicit and specific 'Use when...' trigger guidance, named tool examples, and a tightly scoped ALFWorld niche. The only minor weaknesses are a few missing natural trigger synonyms and a slight gap in covering the full action surface (taking the tool once found).

DimensionReasoningScore

Specificity

The description names concrete actions — 'Searches for a specified tool or device... by checking relevant surfaces' and 'navigates to likely storage spots (sidetables, shelves, countertops)' — with concrete tool examples (desklamp, knife, sponge) in third-person voice. It lists several specific actions but has minor gaps (e.g., it never mentions taking/picking up the tool once found), which keeps it just below the comprehensive-coverage anchor of 5.

4 / 5

Completeness

It explicitly answers both: what — 'Searches for a specified tool or device within the ALFWorld environment by checking relevant surfaces... navigates to likely storage spots... until the tool is found' — and when — 'Use when you need a tool to interact with another object as part of a task but the tool is not in your inventory or immediate vicinity.' The 'when' clause is explicit and concrete rather than merely present, matching the 5 anchor.

5 / 5

Trigger Term Quality

It includes natural trigger terms users would say — 'tool', 'desklamp', 'knife', 'sponge', 'find... not in your inventory' — plus environment-specific phrasing ('ALFWorld', 'sidetables, shelves, countertops'). Coverage is good but a few natural variations are missing (e.g., 'locate', 'where is', 'search for'), so it sits between the good (4) and comprehensive (5) anchors, closer to 4.

4 / 5

Distinctiveness Conflict Risk

The scope is tightly bound to 'the ALFWorld environment' and tool-location triggers (specific object names, receptacles), forming a clear niche with distinct trigger phrases and minimal overlap risk with other skills. Nothing generic like 'helps with tools' remains that would fire the wrong skill.

5 / 5

Total

18

/

20

Passed

Validation

100%

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
Reviewed

Table of Contents

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.