CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

web-search

Real-time web search using Playwright-controlled browser. Use this skill when you need current information, latest documentation, recent news, or any data beyond your knowledge cutoff (January 2025).

Quality

72%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Critical

Do not install without reviewing

Fix and improve this skill with Tessl

tessl review fix ./SKILLs/web-search/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

56%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 highly actionable with executable commands, real error messages, and useful recovery procedures, but it is roughly twice as long as it needs to be due to heavily duplicated example sections. It also claims reference files (README.md, examples/basic-search.md) that are absent from the bundle, and uses inconsistent path conventions ($SKILLS_ROOT vs SKILLs/).

Suggestions

Merge the near-duplicate 'Common Patterns' and 'Examples for Claude' sections into one, and consolidate the overlapping 'Best Practices' and 'Tips for Effective Use' lists to cut the body roughly in half.

Create the referenced README.md and examples/basic-search.md (or remove the pointers), and move the Direct API curl reference and full troubleshooting tables into those files so SKILL.md stays a lean overview.

Standardize all command paths on the documented `$SKILLS_ROOT/web-search/scripts/...` convention — several sections use non-executable `SKILLs/...` variants that will fail as written.

DimensionReasoningScore

Conciseness

The ~550-line body contains substantial padding: "Common Patterns" (5 patterns) and "Examples for Claude" (5 examples) are near-duplicate command listings, "When to Use This Skill" restates the frontmatter description, and "Best Practices"/"Tips for Effective Use" overlap heavily. It avoids explaining generic concepts Claude already knows, so it is not a 1.

2 / 5

Actionability

Most guidance is copy-paste executable (`bash "$SKILLS_ROOT/web-search/scripts/search.sh" "query" 5`, concrete curl calls with endpoint and payload, exact error strings with fixes), but later sections switch to non-executable `SKILLs/web-search/scripts/...` paths that do not match the documented `$SKILLS_ROOT` convention, leaving minor gaps.

4 / 5

Workflow Clarity

The main workflow is clearly sequenced (search, parse, follow-up) and the error-handling and troubleshooting sections provide real feedback loops (stop server, clear .connection cache, restart, test). It falls short of 5 because main-workflow steps like "Analyze results and answer user" have no embedded checkpoints — validation guidance lives only in the separate error section.

4 / 5

Progressive Disclosure

Sections are clearly headed, but ~550 lines of content that belongs in separate files (full curl API reference, duplicated example sets, extensive troubleshooting tables) are inlined, and "Additional Resources" points to `SKILLs/web-search/README.md` and `examples/basic-search.md` which do not exist in the bundle (only scripts/ is present). References exist but are partly broken and the split is weak, matching the some-structure anchor.

3 / 5

Total

13

/

20

Passed

Description

78%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 well-constructed: it states a clear capability, gives an explicit "Use when" clause with concrete trigger phrases tied to the knowledge cutoff, and stays concise. Its only weaknesses are limited action coverage (just "web search") and a few missing natural synonyms.

DimensionReasoningScore

Specificity

"Real-time web search using Playwright-controlled browser" names one concrete action with a concrete mechanism, matching the 1-2-actions anchor; it does not list several distinct actions, so it is not a 4, and it is more concrete than the generic 'Processes PDF files' level.

3 / 5

Completeness

It explicitly answers both what ("Real-time web search using Playwright-controlled browser") and when ("Use this skill when you need current information, latest documentation, recent news, or any data beyond your knowledge cutoff") with concrete trigger phrases, mirroring the anchor-5 example.

5 / 5

Trigger Term Quality

"current information", "latest documentation", "recent news", and "knowledge cutoff" are natural user phrases with good coverage, but common variations like "search online" or "look this up" are missing, keeping it below the comprehensive anchor.

4 / 5

Distinctiveness Conflict Risk

Web search via a Playwright-controlled browser is a clear, mostly distinct niche with specific triggers; minor overlap risk remains with a general fetch/browse skill on phrases like "latest documentation", so it is not a 5.

4 / 5

Total

16

/

20

Passed

Validation

87%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

skill_md_line_count

SKILL.md is long (562 lines); consider splitting into references/ and linking

Warning

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

14

/

16

Passed

Repository

netease-youdao/LobsterAI

Path

SKILLs/web-search/SKILL.md

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.