scienceworld-recipe-retriever

This skill locates and acquires a recipe or instruction document by using 'pick up' on the recipe object. Use when the task involves following a specific procedure (e.g., crafting, mixing) and the agent needs to obtain the written instructions. The skill assumes the recipe is visible in the current room and moves it to the inventory, allowing the agent to read it later to understand required ingredients and steps.

Quality

90%

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

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.

The content is lean, actionable, and has a clear workflow with validation and error feedback. The main gap is progressive disclosure: existing reference bundle files are unreferenced and partly duplicated inline instead of being linked one level deep.

Suggestions

Replace the inline action/error detail in 'Core Procedure' and 'Error Handling' with one-level-deep links to the existing bundle files, e.g. 'See [references/action_primitives.md](references/action_primitives.md) for the full action set and observation formats.'

Link references/trajectory_analysis.md from a short 'How this skill fits a larger task' note so the integration guidance is referenced rather than re-derived inline.

Keep only the minimal command set in the body and move the enumerated observation patterns / error-message catalog into the references, signaling them clearly to avoid duplicated content.

Dimension	Reasoning	Score
Conciseness	Lean ~32-line body that assumes Claude's competence — no padding about what recipes are or how the environment works; every section (Purpose, Triggers, Procedure, Notes, Errors, Example) earns its tokens.	5 / 5
Actionability	Concrete, copy-paste-ready commands (`look around`, `pick up <recipe_object>`, `read`) plus a worked example with exact inputs and expected observations, fully executable for the common case.	5 / 5
Workflow Clarity	A clearly sequenced Locate → Acquire → Verify procedure with an explicit verification checkpoint ("Confirm the recipe is now in your inventory before proceeding") and an error-recovery feedback loop (pick-up fails → examine; not found → search adjacent rooms).	5 / 5
Progressive Disclosure	The body is well-sectioned, but two real bundle files (references/action_primitives.md and trajectory_analysis.md) are never referenced from the body, and their content (action primitives, observation patterns, error messages) is partly duplicated inline rather than linked — references exist but are not signaled, fitting the score-3 anchor.	3 / 5
	Total	18 / 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.

The description is specific, complete, and distinct: it states concrete actions, an explicit "Use when..." trigger, and a narrowly scoped niche. The only mild weakness is trigger-term breadth (a few synonyms/extensions missing).

Dimension	Reasoning	Score
Specificity	Names the domain and several concrete actions — "locates and acquires a recipe or instruction document by using 'pick up' on the recipe object" and "moves it to the inventory, allowing the agent to read it later" — with only minor coverage gaps, fitting the score-4 anchor rather than the less comprehensive 3.	4 / 5
Completeness	Explicitly answers both what (locate and acquire the recipe via pick up, move to inventory) and when ("Use when the task involves following a specific procedure... and the agent needs to obtain the written instructions"), matching the score-5 anchor.	5 / 5
Trigger Term Quality	Good natural-term coverage ("recipe", "instruction document", "crafting, mixing") but missing common synonyms like "manual"/"blueprint" and any file-like extensions, so it sits above 3 yet below the comprehensive 5.	4 / 5
Distinctiveness Conflict Risk	Carves a clear niche — recipe/instruction document retrieval via a specific pick-up action — with distinct triggers and minimal overlap risk with other skills.	5 / 5
	Total	18 / 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-recipe-retriever/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.