CtrlK
BlogDocsLog inGet started
Tessl Logo

scienceworld-substance-fetcher

Locates and retrieves a target substance or material from a container in the environment. Use this skill when the task requires processing a specific substance (e.g., chocolate, sodium chloride, a chemical) and you need to find and acquire it. Searches rooms and containers (fridge, cupboard, counter) and retrieves the substance via pick up or move actions.

65

Quality

82%

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

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 concise, actionable, single-purpose skill with a clear workflow and error-recovery loops, appropriate for a short SKILL.md. Its main weakness is progressive disclosure: the bundled references/action_primer.md is orphaned — never linked from the body — while parts of its content (container names, room mappings, action syntax) are duplicated inline.

Suggestions

Reference the bundle explicitly, e.g. under Search Strategy: 'Full action syntax and container/room mappings: see [action_primer.md](references/action_primer.md)', and drop the inline duplication of container names and room-substance mappings in favor of that pointer.

State the container-open constraint once (Critical Constraints) instead of twice, and delete the duplicated 'Key Assumption' line in step 4.

Make the verification checkpoint explicit rather than optional: 'After retrieval, confirm the substance is in your inventory before proceeding' — replace the hedged 'you may use examine or check your inventory'.

DimensionReasoningScore

Conciseness

The body is lean and assumes competence (no explanation of what teleporting or containers are), but has minor redundancy: 'All containers are already open. Do not use open or close actions' appears in step 4 and is restated in Critical Constraints, and the Example Execution re-derives steps already given. Not 5 due to this duplication; not 3 because nearly every token earns its place.

4 / 5

Actionability

Provides concrete commands — 'teleport to [room]', 'look around', 'pick up [substance]', 'move [substance] to [destination container]', 'examine [substance]' — plus a concrete worked example ('teleport to kitchen', 'pick up chocolate'). Not 5 because the general commands use bracketed placeholders rather than copy-paste-ready literals, and edge-case syntax (e.g., multi-word object names) is unspecified; not 3 because the guidance is directly executable in the target environment.

4 / 5

Workflow Clarity

A clear numbered sequence (identify target → search → locate in container → retrieve) with an Error Handling section giving recovery loops ('teleport to and search other relevant rooms', 're-examine the room description with look around'). Not 5 because the verification checkpoint is hedged and optional — 'you may use examine [substance] or check your inventory' — rather than an explicit validate-before-proceeding step; not 3 because sequence, checkpoints, and recovery paths are all present.

4 / 5

Progressive Disclosure

The body itself is well-organized under clear section headers, but the bundle contains references/action_primer.md (action syntax, container names, room-to-substance mappings) that is never referenced or linked anywhere in the body. Per the guideline to score against the actual bundle structure, this is 'references present but not clearly signaled' — the body even duplicates a subset of the primer's content inline. Not 4 because an existing complementary bundle file has zero navigation to it; not 2 because the inline content is structured and the skill functions standalone.

3 / 5

Total

15

/

20

Passed

Description

83%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 that explicitly states both what the skill does and when to use it, with concrete examples and named actions. Main gaps are missing synonym coverage for trigger terms and the absence of the environment name, which slightly limits distinctiveness.

DimensionReasoningScore

Specificity

Quotes several concrete actions — 'Locates and retrieves a target substance', 'Searches rooms and containers (fridge, cupboard, counter)', 'retrieves the substance via pick up or move actions' — with only minor coverage gaps. Not 5 because the action list is somewhat repetitive around locate/retrieve and offers little beyond the fetch loop; not 3 because multiple specific actions plus named containers and named commands are given.

4 / 5

Completeness

Explicitly answers both: what — 'Locates and retrieves a target substance or material from a container... via pick up or move actions' — and when — 'Use this skill when the task requires processing a specific substance (e.g., chocolate, sodium chloride, a chemical) and you need to find and acquire it', with concrete trigger phrases. Not 4 because the 'when' clause is explicit and specific rather than generic.

5 / 5

Trigger Term Quality

Includes natural terms users would say: 'substance', 'material', 'chemical', concrete examples 'chocolate', 'sodium chloride', and the need phrase 'find and acquire it'. Not 5 because common synonyms like 'ingredient', 'compound', or 'reagent' are missing; not 3 because several natural, varied trigger terms with concrete examples are present.

4 / 5

Distinctiveness Conflict Risk

The substance-fetching niche with named containers and actions is mostly distinct from generic manipulation skills. Not 5 because the description never names the target environment (ScienceWorld), leaving 'the environment' somewhat generic within a family of environment skills; not 3 because the substance-acquisition focus with concrete containers is clearly distinguishable.

4 / 5

Total

17

/

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.