CtrlK
BlogDocsLog inGet started
Tessl Logo

scienceworld-ambiguous-action-resolution

Use when the ScienceWorld environment returns an "Ambiguous request" prompt with a numbered list of identical action options. This skill resolves the disambiguation by selecting the lowest available number (typically 0) to proceed, ensuring task progression when multiple identical object instances exist and the system cannot automatically determine which instance to act upon.

68

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

77%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 delivers fully executable, example-backed guidance with a clear workflow for a simple skill, and its structure is easy to navigate. The weaknesses are cross-section redundancy (When to Use / Key Principles / Anti-Patterns repeat the same caveats, and the two examples are near-duplicates) and an orphaned, truncated references/trajectory_examples.md that the body never points to.

Suggestions

Cut redundancy: merge the Key Principles section into Workflow/Anti-Patterns (the identical-instances caveat currently appears in three places) and drop or differentiate Example 2, which teaches the same select-0 lesson as Example 1.

Fix the bundle reference: either link references/trajectory_examples.md from the body (e.g., under When to Use) and complete its truncated content, or remove the file — currently it is both unreachable and cut off mid-sentence.

Tighten 'Efficiency -- Do not overthink' and similar advisory phrasing; the workflow already implies it, so these tokens earn little.

DimensionReasoningScore

Conciseness

The body is mostly efficient but includes tightening opportunities: "Efficiency -- Do not overthink; the objects are identical and any choice is valid" restates the workflow, the Key Principles section repeats When to Use and Anti-Patterns content (e.g., the identical-instances point appears in three sections), and Example 2 teaches the same lesson as Example 1 with a different surface. Not anchor 2 because there is no padding or explanation of concepts Claude already knows; not anchor 4 because the redundancy across sections is more than minor.

3 / 5

Actionability

The guidance is fully executable for an instruction-only skill: it quotes the exact trigger phrase ("Ambiguous request: Please enter the number for the action you intended (or blank to cancel):"), prescribes the exact action ("Choose `0`"), and provides two complete observation-to-action examples that are copy-paste ready against the common cases. Not below 5 because no key detail is missing.

5 / 5

Workflow Clarity

This is a simple single-purpose skill with an unambiguous four-step sequence (Recognize the prompt, Scan the options, Select the lowest number, Execute) that includes an inline checkpoint ("Confirm that the listed options (0, 1, 2...) are functionally identical") plus error-path guidance in Anti-Patterns ("Do NOT trigger if the observation does not contain the exact 'Ambiguous request' phrase"). The operation is neither destructive nor batch, so the validation cap does not apply, and the simple-skill exception holds. Not below 5 because no step is ambiguous and failure modes are explicitly covered.

5 / 5

Progressive Disclosure

The body itself is well-sectioned (Purpose, When to Use, Workflow, Examples, Key Principles, Anti-Patterns), but the bundle contains references/trajectory_examples.md — which is truncated mid-sentence ("The ambiguity prompt always follows this structure:") — and the body never links to or signals it, so the reference is present but not clearly signaled. Not anchor 4 because an unreferenced and incomplete bundle file is more than a minor organization gap; not anchor 2 because the inline content is appropriately sized and well structured.

3 / 5

Total

16

/

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.

A strong description: it explicitly states what the skill does and when to use it, anchored by the exact environment prompt phrase that serves as a natural trigger. The only weakness is thin synonym coverage around the trigger terms.

DimensionReasoningScore

Specificity

The description names the domain (ScienceWorld disambiguation prompts) and concrete actions — "resolves the disambiguation by selecting the lowest available number (typically 0) to proceed, ensuring task progression". It falls between anchor 3 (1-2 concrete actions) and anchor 5 (multiple actions, comprehensive), and since the single action is fully specified for this single-purpose skill but there are not several distinct actions, 4 is the best fit.

4 / 5

Completeness

It clearly answers both: what ("resolves the disambiguation by selecting the lowest available number (typically 0) to proceed") and when ("Use when the ScienceWorld environment returns an 'Ambiguous request' prompt with a numbered list of identical action options") with a concrete trigger phrase, matching the anchor-5 example structure exactly. It is not anchor 4 because the 'when' clause is explicit and specific rather than improvable.

5 / 5

Trigger Term Quality

Good keyword coverage including the exact trigger phrase users/observations would contain — "Ambiguous request" prompt, "numbered list of identical action options", "ScienceWorld environment", "identical object instances". It lacks synonym or variation coverage (e.g., disambiguation prompt, instance selection), so it does not reach the comprehensive anchor 5.

4 / 5

Distinctiveness Conflict Risk

It occupies a clear niche — disambiguation of identical object instances in the ScienceWorld environment — with a distinct trigger phrase ("Ambiguous request") unlikely to fire for any other skill, so conflict risk is minimal.

5 / 5

Total

18

/

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.