CtrlK
BlogDocsLog inGet started
Tessl Logo

scienceworld-container-inspector

Inspects the contents of a container or device using the 'look at' action. Use this skill when you need to verify what is inside a container (e.g., checking if lead is in the blast furnace), monitor the state of contents (e.g., solid vs. liquid phase changes), or confirm that a placement or process step succeeded. Returns a detailed list of contents and their states for process monitoring.

72

Quality

90%

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

92%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.

An exemplary instruction-only skill body: exact command syntax, a worked example with interpretation, a placement-in-sequence diagram, and concrete error-recovery guidance, all in a lean 42 lines. The sole real defect is progressive disclosure — the bundled references/container_inspection_examples.md is never referenced from the body and is itself incomplete, leaving a dead, undiscoverable file.

Suggestions

Reference the bundle file explicitly, e.g. under a 'More examples' section: "See [references/container_inspection_examples.md](references/container_inspection_examples.md) for additional observation patterns" — or remove it if the inline trajectory example is sufficient.

Complete or prune references/container_inspection_examples.md: it currently cuts off mid-example after "**Action:** `look at cupboard` / **Observation:**" with no content, which is worse than no reference at all once discovered.

Consider moving the 'Example from Trajectory' section into the references file (linked from the body) if more examples are added, keeping SKILL.md focused on the command syntax and interpretation rules.

DimensionReasoningScore

Conciseness

The ~42-line body is lean with zero padding: every section carries domain-specific knowledge Claude would not have (the observation pattern, interpretation rules, a real trajectory example, error handling). It does not over-explain known concepts, matching anchor 5's 'every token earns its place' rather than anchor 4's 'minor instances of over-explanation'.

5 / 5

Actionability

"Always use the exact action: `look at <CONTAINER_NAME>`" plus a fully worked example (`look at blast furnace` with its observation and interpretation) and concrete error-recovery commands (`look around`, `examine`, `open`) make the guidance copy-paste executable in the target environment. This matches anchor 5; it is not anchor 4 because there are no gaps in the guidance for the common cases.

5 / 5

Workflow Clarity

This is a simple single-purpose skill whose single action is unambiguous, which per the simple-skill guideline can score 5. It additionally provides a numbered integration sequence showing where the action fits and an explicit error-recovery loop (verify object name via 'look around'/'examine'; open a closed door before inspecting), so it sits at anchor 5 rather than 4. No destructive or batch operations are involved, so no cap applies.

5 / 5

Progressive Disclosure

The body itself is well-organized with clear sections, but the bundle contains references/container_inspection_examples.md which is never mentioned or linked anywhere in the body — a reference present but not clearly signaled (and the file itself is truncated mid-example). This matches anchor 3; it is above anchor 2 because the inline content is appropriately scoped and sectioned, and below anchor 4 because the one existing bundle file is completely un-signaled and orphaned.

3 / 5

Total

18

/

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 the concrete action, mechanism, and return value, and pairs them with a well-formed 'Use this skill when...' clause covering three distinct use cases with concrete examples. Written in third person with no fluff or over-claims; only minor gains available from broader synonym coverage.

DimensionReasoningScore

Specificity

"Inspects the contents of a container or device using the 'look at' action" and "Returns a detailed list of contents and their states" name the concrete mechanism and output with concrete examples (blast furnace, lead). It sits above anchor 3 because more than 1-2 actions/outputs are specified, but below anchor 5 since the capability is essentially one inspect action with use-case variants rather than multiple distinct concrete actions.

4 / 5

Completeness

Both questions are explicitly answered: "Inspects the contents of a container or device using the 'look at' action" states what it does, and "Use this skill when you need to verify... monitor... or confirm..." gives three explicit trigger scenarios with concrete examples. This clearly matches anchor 5 rather than 4, since the 'when' clause is fully explicit with concrete trigger phrases rather than improvable.

5 / 5

Trigger Term Quality

Natural phrases like "verify what is inside a container", "monitor the state of contents", "solid vs. liquid phase changes", and "checking if lead is in the blast furnace" give good keyword coverage with verify/monitor/confirm/check variations. A few natural terms are missing (e.g., "examine", "inventory", "check contents"), keeping it below anchor 5's comprehensive synonym coverage.

4 / 5

Distinctiveness Conflict Risk

The description carves a clear niche ('look at' action for container/device contents, phase-change monitoring, process verification) with distinct triggers like the blast furnace example. Broad words such as "container" and "device" leave minor overlap risk with other inspection-type skills, so it does not reach anchor 5.

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.