searxng-search

Enhanced web and package repository search using local SearXNG instance

3.80x

Quality

52%

Does it follow best practices?

Impact

95%

3.80x

Average score across 3 eval scenarios

Securityby

Advisory

Suggest reviewing before use

Optimize this skill with Tessl

npx tessl skill review --optimize ./searxng-search/SKILL.md

Evaluation results

100%

82%

Async Runtime Crate Research Tool

Rust crate search workflow and category selection

Criteria

Without context

With context

Checks if running first

100%

Uses start-searxng helper

100%

Correct base URL

100%

format=json in URL

100%

Uses cargo category

100%

Fallback to packages or it

100%

Parses with jq

100%

Limits result count

100%

Stops SearXNG after use

100%

Outputs results.json

100%

88%

63%

HTTP Client Library Survey

PyPI workaround and multi-ecosystem package search

Criteria

Without context

With context

Rust uses cargo category

100%

JS uses packages category

100%

No SearXNG PyPI for Python

100%

PyPI JSON API or qypi

100%

format=json in SearXNG URLs

100%

Checks/starts SearXNG

50%

Filters by engine for npm

100%

Parses with jq

100%

Stops SearXNG after use

Output organized by ecosystem

100%

99%

65%

Transformer Architecture Research Aggregator

Multi-category research workflow with parallel searches

Criteria

Without context

With context

scientific publications category

91%

repos category

100%

it category for articles

100%

Parallel background searches

100%

format=json in all URLs

100%

Checks/starts SearXNG

100%

Uses start-searxng helper

100%

Limits results to ≤5 per section

100%

Stops SearXNG after use

100%

Output has distinct sections

100%

Configurable topic variable

100%

Repository: YPares/agent-skills
Commit: aed1afb

Evaluated: 29 days ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Async Runtime Crate Research Tool HTTP Client Library Survey Transformer Architecture Research Aggregator

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.