alfworld-navigation-planner

Plans a path to move the agent between receptacles to search for target objects. Use this when you need to traverse the environment to reach a specific location or systematically explore multiple areas. It takes the current location and destination receptacle as input, and outputs the next 'go to' action to approach the target.

Quality

78%

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

Fix and improve this skill with Tessl

tessl review fix ./experiments/src/skills/alfworld/alfworld-navigation-planner/SKILL.md

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 concise and gives a clear input/output contract with a usable worked example, but it is not fully actionable: it defers core logic to a bundled script that is neither present nor invocable as described, and it fails to link the one reference file that does exist. Workflow sequencing is good, with only vague failure-recovery guidance keeping it from the top anchor.

Suggestions

Add concrete invocation guidance: name the actual script path and how to call it with the current location and target receptacle, or remove the 'core logic is handled by the bundled script' claim since no scripts/ file exists in the bundle.

Link the existing reference file, e.g. 'See [action_primer.md](references/action_primer.md) for the full action set, failure semantics, and receptacle-to-object associations', so the bundled material is discoverable.

Replace 'the script's logic helps choose an alternative path' with an explicit recovery step the agent can execute when a 'go to' action fails, to strengthen the workflow's feedback loop.

Dimension	Reasoning	Score
Conciseness	The body is lean with well-organized sections and no explanation of concepts Claude already knows; only the opening line and parts of 'Key Principles' restate the description, which could be trimmed. Close to but not quite the 'every token earns its place' anchor.	4 / 5
Actionability	The input/output format and the worked example ('desk 1', 'bed 1', desklamp) are concrete, but key execution details are missing: the body says 'The core logic is handled by the bundled script' yet no script path or invocation instruction is given (and no scripts/ file exists). Better than 2 — the I/O contract and example are specific — but incomplete like the anchor's 'missing key details'.	3 / 5
Workflow Clarity	The four-step 'How to Use' sequence is clear with an explicit observe-and-repeat loop and failure handling ('If an action fails ("Nothing happened")...'). Not a 5 because error recovery is delegated vaguely ('the script's logic helps choose an alternative path') rather than given as a concrete checkpoint the agent can execute.	4 / 5
Progressive Disclosure	Sections are well organized for a short skill, but scored against the actual bundle: references/action_primer.md exists and is never referenced or linked from the body, and the claimed 'bundled script' does not exist in the bundle. This matches 'references present but not clearly signaled' rather than the good-structure anchor.	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.

A strong description: it uses third-person voice, defines a concrete input/output contract, and pairs a clear capability statement with an explicit, multi-scenario 'Use when' clause. Its only weaknesses are a slightly generic 'search for target objects' phrasing and a few missing natural synonyms like 'navigate to'.

Dimension	Reasoning	Score
Specificity	Names several concrete actions in third person ('Plans a path to move the agent between receptacles', 'outputs the next 'go to' action') with a defined input/output contract. Not a 5 because coverage is not comprehensive — 'search for target objects' stays generic and no other actions or edge behaviors are enumerated.	4 / 5
Completeness	Explicitly answers both: what ('Plans a path... outputs the next 'go to' action to approach the target' with inputs specified) and when ('Use this when you need to traverse the environment to reach a specific location or systematically explore multiple areas'). The when-clause contains two concrete trigger scenarios, matching the top anchor.	5 / 5
Trigger Term Quality	Includes natural trigger phrases like 'traverse the environment', 'reach a specific location', 'systematically explore multiple areas', and the literal 'go to' action. Not a 5 because common synonyms a user would naturally say ('navigate to', 'move to', 'find objects') are missing; better than 3 since coverage is genuinely good rather than partial.	4 / 5
Distinctiveness Conflict Risk	The receptacle-navigation niche and 'go to' action framing are mostly distinct from other skills. Not a 5 because 'systematically explore multiple areas' and 'search for target objects' are broad enough to overlap with generic exploration or object-search skills.	4 / 5
	Total	17 / 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/alfworld/alfworld-navigation-planner/SKILL.md
Commit: 358ec82

Reviewed: 8 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.