search-fallback-workflow

Graceful degradation workflow for continuing tasks when web search tools fail, using internal knowledge with verification caveats

Quality

50%

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 ./benchmarks/gdpval/skills/search-fallback-workflow/SKILL.md

Quality

Content

50%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The skill is well-structured with a clear workflow and useful output templates, but it relies on pseudocode rather than executable code, repeats logic across sections, and keeps all material inline with no progressive file structure.

Suggestions

Replace the pseudocode block with runnable, complete code (or explicitly justify why pseudocode is sufficient) to lift actionability.

Consolidate the duplicated detection logic between the Core Pattern and Code Pattern sections to remove redundancy and tighten conciseness.

Move the Implementation Template and Example Output Structure into a referenced file (e.g. references/templates.md) to introduce one-level-deep progressive disclosure.

Dimension	Reasoning	Score
Conciseness	The body avoids explaining concepts Claude already knows, but the Core Pattern steps and the "Code Pattern for Agents" repeat the same detection logic, and sections like Implementation Template, Code Pattern, and Example Output Structure overlap, so it could be tightened beyond the score-3 lean ideal.	2 / 3
Actionability	The markdown templates (verification checklist, execution notes) are concrete and copy-paste ready, but the core logic is explicitly labeled "# Pseudocode for search fallback detection" rather than executable code, matching the score-2 "pseudocode instead of executable code" anchor.	2 / 3
Workflow Clarity	The five-step Core Pattern is clearly sequenced and includes an assessment checkpoint and a verification checklist, but there is no validate-then-fix-then-retry feedback loop, leaving checkpoints implicit rather than the explicit validation loops expected for a score 3.	2 / 3
Progressive Disclosure	The file is well-organized into clear sections, but at roughly 130 lines everything is inline with no bundle files or one-level-deep references, so content that could be split (templates, code pattern) remains monolithic rather than the score-3 split-with-navigation structure.	2 / 3
	Total	8 / 12 Passed

Description

50%

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 conveys both purpose and a triggering condition concisely, but it leans on abstract process language, omits an explicit "Use when..." clause, and risks overlapping with general error-recovery skills.

Suggestions

Add an explicit trigger clause, e.g. "Use when `search_web` repeatedly fails or returns empty results and the task can still be completed with internal knowledge."

Replace abstract phrasing ("graceful degradation workflow") with concrete actions the skill performs, such as "Detect search failures, continue with internal knowledge, and flag unverified claims for follow-up."

Include natural trigger variations users would say ("search unavailable", "search not working", "search timed out") to improve trigger-term coverage.

Dimension	Reasoning	Score
Specificity	Phrases like "Graceful degradation workflow", "continuing tasks", and "using internal knowledge with verification caveats" name the domain and some actions, but they describe a process abstractly rather than listing multiple concrete actions, so it sits below the score-3 "lists multiple specific concrete actions" anchor.	2 / 3
Completeness	The what ("graceful degradation workflow...") and a when ("when web search tools fail") are both present, but there is no explicit "Use when..." clause, which the guidelines say caps completeness at 2 rather than 3.	2 / 3
Trigger Term Quality	"when web search tools fail" is a relevant trigger phrase a user might echo, but common variations ("search unavailable", "search not working", "search broken") are absent, matching the score-2 "some relevant keywords but missing common variations" anchor.	2 / 3
Distinctiveness Conflict Risk	"Graceful degradation workflow" is a fairly generic meta-process label and the body itself lists overlapping skills like `error-recovery-workflow`, so it could overlap with similar recovery patterns rather than occupying a clearly distinct niche.	2 / 3
	Total	8 / 12 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/search-fallback-workflow/SKILL.md
Commit: c5a9c4b

Reviewed: 2 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.