web-tool-fallback

Implement fallback strategies when web-reading tools fail simultaneously

Quality

44%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Critical

Do not install without reviewing

Fix and improve this skill with Tessl

tessl review fix ./benchmarks/gdpval/skills/web-tool-fallback/SKILL.md

Quality

Content

42%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 skill delivers a workable fallback procedure with executable snippets and a decision tree, but suffers from duplicated section headers, a broken/unterminated code fence, and verbosity that reduce clarity and actionability. Fixing the malformed fence and de-duplicating headers would notably improve it.

Suggestions

Remove the duplicated headers for Step 2 (lines 47-48) and Step 2b (lines 81-82), keeping only one of each.

Fix the broken code fence around lines 75-80 where a ```python block is never closed and bullet list items are rendered as Python source.

Trim verbose justification sections (e.g., 'Why run_shell works when execute_code_sandbox fails') and consolidate the checklist, decision tree, and Best Practices so each token earns its place.

Dimension	Reasoning	Score
Conciseness	The body is noticeably verbose with redundant duplicated headers (Step 2 and Step 2b appear twice), a checklist and decision tree that restate the procedure, and explanatory lines like 'Why run_shell works...' that pad the content without adding novel instruction.	2 / 5
Actionability	It provides concrete runnable bash/python snippets for the fallbacks, but a broken code fence (lines 75-80 leave a Python block unterminated and mix bullets into code) and a non-executable example Application make the guidance only partially executable.	3 / 5
Workflow Clarity	The four-step procedure is sequenced with a decision tree, but the duplicated Step 2/2b headers and a broken code fence break the flow; while error-then-fallback checkpoints exist, the malformed structure introduces gaps and ambiguity.	3 / 5
Progressive Disclosure	There are no bundle files (references/scripts/assets absent) and the content is organized into labeled sections with a decision tree and adaptations table, but a fair amount of inline code/templates that could live in reference files keeps it only moderately organized.	3 / 5
	Total	11 / 20 Passed

Description

46%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 reasonably distinct and niche-specific but lacks a 'Use when...' trigger clause and natural user-facing trigger terms. Adding concrete fallback actions and explicit usage triggers would lift the weaker dimensions.

Suggestions

Add a 'Use when...' clause with concrete triggers, e.g. 'Use when all web-reading tools (read_webpage, search_web) fail simultaneously with unknown errors during research.'

Enumerate the concrete fallback actions (e.g., try archive.org URLs, generate content from embedded knowledge, use run_shell) instead of the generic 'implement fallback strategies'.

Include natural user phrasings like 'web tools failed', 'cannot access web', or 'all web tools down' to improve trigger-term quality.

Dimension	Reasoning	Score
Specificity	Description names the domain (web-reading tool fallback) and one concrete action ('implement fallback strategies') but does not enumerate the specific fallback actions or tools involved, leaving coverage incomplete.	3 / 5
Completeness	It gives a clear 'what' (implement fallback strategies) but provides no 'Use when...' clause or equivalent explicit trigger guidance, capping completeness at 3 per the rubric.	3 / 5
Trigger Term Quality	It contains generic phrases like 'fallback strategies' and 'web-reading tools fail' but lacks the natural trigger terms (e.g., 'web tools failed', 'cannot access web', 'all web tools down') a user would actually say, missing common variations.	2 / 5
Distinctiveness Conflict Risk	The 'when all web-reading tools fail simultaneously' niche is fairly distinct and unlikely to trigger for the wrong skill, with only minor overlap risk against general research skills.	4 / 5
	Total	12 / 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: HKUDS/OpenSpace
Path: benchmarks/gdpval/skills/web-tool-fallback/SKILL.md
Commit: 3827781

Reviewed: 5 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.