CtrlK
BlogDocsLog inGet started
Tessl Logo

scienceworld-item-fetcher

Picks up a specified object from the environment and moves it into the agent's inventory. Use this skill when you identify a required tool or material (e.g., thermometer, metal pot) that needs to be collected for later use in an experiment or task. Takes an object name as input and confirms the item's transfer to inventory.

62

Quality

73%

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

Fix and improve this skill with Tessl

tessl review fix ./experiments/src/skills/scienceworld/scienceworld-item-fetcher/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

63%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.

The body is a clear, actionable, well-sequenced guide for a simple text-game skill, with a verification step and concrete commands. Its weaknesses are redundant restating of the workflow across sections and an unlinked reference file that breaks progressive disclosure.

Suggestions

Link `references/action_primer.md` from the body (e.g., a "See action_primer.md for supporting actions and pitfalls" line) so the bundled detail is discoverable.

Collapse the redundancy between Core Workflow, Key Principles, and Example from Trajectory — keep one canonical sequence and let the example illustrate only the novel observation/result strings.

Add the concrete `open OBJ` command to the Error Handling step so the container-failure recovery is fully executable.

DimensionReasoningScore

Conciseness

Mostly efficient but the Key Principles and Example from Trajectory sections restate the Core Workflow (look around, pick up, container-error handling), creating noticeable redundancy. Not a 4 because the repetition is more than minor trimming; not a 2 because the prose itself is not padded with concepts Claude already knows.

3 / 5

Actionability

Provides concrete executable commands — `look around`, `pick up OBJ`, `focus on OBJ` — plus a worked trajectory example (`pick up thermometer`). Not a 5 because the error-handling guidance ("Ensure all relevant containers are open first") names no concrete `open` command, leaving a small execution gap.

4 / 5

Workflow Clarity

A clear 4-step sequence (Identify Need → Locate via `look around` → Execute `pick up` → Verify Success) with an explicit verification checkpoint and a brief error-recovery note. Not a 5 because the feedback loop for the container-failure case is light and does not fully re-validate after recovery.

4 / 5

Progressive Disclosure

The body is cleanly sectioned, but a bundle file `references/action_primer.md` exists and is never referenced or signaled from the overview, so useful detail (teleport to, look at, the "all containers already opened" assumption) is unreachable by navigation. Not a 4 because the missing signal to a real reference file is more than a minor gap.

3 / 5

Total

14

/

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.

The description is strong: third-person, concise, and explicitly covers both what the skill does and when to invoke it with concrete trigger examples. The main limitation is modest trigger-term breadth and some residual overlap with related object-handling skills.

DimensionReasoningScore

Specificity

Names concrete actions — "Picks up a specified object... and moves it into the agent's inventory", "confirms the item's transfer" — covering several specific actions for a narrow task. Not a 5 because the action set is small and tailored to a single-purpose skill rather than comprehensive coverage of multiple capabilities.

4 / 5

Completeness

Explicitly answers both "what" (picks up, moves to inventory, takes object name, confirms transfer) and "when" ("Use this skill when you identify a required tool or material... that needs to be collected") with concrete trigger phrases. Clearly satisfies the top anchor; the "Use when" clause is present so it is not capped at 3.

5 / 5

Trigger Term Quality

Includes natural trigger terms and concrete examples — "required tool or material (e.g., thermometer, metal pot)", "collected for later use in an experiment or task". Not a 5 because it lacks broader synonym/variation coverage beyond the two named examples.

4 / 5

Distinctiveness Conflict Risk

Targets a clear niche (fetching a named object into inventory for ScienceWorld experiments) with concrete examples that reduce misfiring. Not a 5 because "picks up an object" could still overlap with other object-manipulation or experiment-setup skills.

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.

Validation16 / 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.