SKILLEMALL.ai
GitHub ↗ RU Sign in

BC جاك العلم

Personal Shopper — multi-agent product/service research and recommendation for Saudi Arabia. USE WHEN: - User asks to find, compare, recommend, or buy a product or service - "what's the best X", "compare X vs Y", "find me a good X" - "أبغى أشتري", "وش أفضل", "قارن لي", "ابحث لي عن" - User asks "is this a good deal" or "should I buy X or Y" - Product comparison by specs, price, or value DON'T USE WHEN: - Market analysis for business entry → use mckinsey-research - Comparing companies as businesses (not products) → use mckinsey-research - Price tracking over time or deal alerts → not supported - Reviewing/troubleshooting a product they already own → answer directly - Simple factual question about a product ("how much RAM does iPhone have") → answer directly - Order placement, returns, or refunds → not supported EDGE CASES: - "أبغى أشتري لابتوب" → this skill - "أبغى أفتح متجر لابتوبات" → mckinsey-research (business, not purchase) - "وش أفضل شاشة" → this skill - "وش حجم سوق الشاشات" → mckinsey-research - "هل السعر هذا حلو على أمازون" → this skill - "حلل لي سوق التجارة الإلكترونية" → mckinsey-research - "قارن لي بين منتجين" → this skill - "قارن لي بين شركتين" → mckinsey-research INPUTS: Product type or name, budget (optional), use case (optional), preferences (optional) TOOLS: sessions_spawn (sub-agents), web_fetch, web_search, camofox_* (with strict limits per agent) OUTPUT: HTML report saved to shopping-reports/{date}-{slug}.html (Arabic, RTL, mobile-friendly) SUCCESS: User gets 3 ranked options with verified prices, source URLs, coupons, and a clear recommendation

modbender/skill-library-mcp Claude Code author: modbender MIT 7 files body ≈ 7 309 tokens Open the sourcegithub.com↗ analyzed 3 d ago

Personal Shopper — multi-agent product/service research and recommendation for Saudi Arabia.

As a process C 51/100 · Has gaps — weak spots: result and completion, inputs and preconditions, consistency

AnalyzerAI and agentstype and topics are labelled automatically from the skill text
Runs in: OpenClaw Claude / Claude Code ChatGPT / Codex Gemini CLI Cursor GitHub Copilot / VS Code Qwen Code Kimi Code CLI Z.ai (GLM) DeepSeek (Deep Code) Mistral Vibe LangChain (deepagents) Goose OpenCode
JSON
Technical rating
B
79/100
safety, quality, tests
Safety 60%
100
Quality 40%
48
Run on models
none yet
Process rating
C
51/100
Has gaps
Result and completion w 14
0
Inputs and preconditions w 11
0
Progress reporting w 2
0
the three weakest of ten parameters · all ten

How to improve

  1. Shorten the description to 1024 characters.
  2. The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
For the model run — optional
  • Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
  • A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.

Guard findings · 0

✓ No critical or high findings

Files scanned: 7. Evidence is masked. Grey chips explain why severity was lowered.

Against the Agent Skills spec

  • error description-long description is 1593 chars, limit 1024
  • warning name-format name should be kebab-case (lowercase letters, digits, hyphens)
  • warning body-long SKILL.md body ≈ 7309 tokens (recommended < 5000); move details to references/
  • note description-budget description takes 1593 of the ~15000-char shared budget for all skills
  • note frontmatter-key unknown frontmatter key "trigger"
  • note frontmatter-key unknown frontmatter key "locale"
  • note frontmatter-key unknown frontmatter key "region"
  • note frontmatter-key unknown frontmatter key "currency"
  • note frontmatter-key unknown frontmatter key "output"
  • note frontmatter-key unknown frontmatter key "output_dir"
  • note frontmatter-key unknown frontmatter key "always"
  • note frontmatter-key unknown frontmatter key "security"

Process rating: all ten parameters 51/100

  • 0Result and completion. Does not say what the result is
  • 0Inputs and preconditions. Does not say what the process needs to start
  • 0Progress reporting. Says nothing while it works
  • 40Consistency. Frontmatter name (جاك العلم) differs from the folder (personal-shopper)
  • 50When it triggers. No condition that starts the skill
  • 60Tools and files. Uses tools (bash, read, web) that frontmatter does not declare
  • 70Execution cost. Instruction body is 7309 tokens
  • 85Steps. 105 steps, 1 vague phrases
  • 100Failures and branches. 1 branches, has a failure section
  • 100Running it twice. Mutating operations check current state
  • medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
  • low 15 top-level sections: this looks like several domains in one skill

Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.

Quality signals

  • +3Description length 1592: 120–800 characters recommended
  • +3Output format is not stated: the model decides each time
  • +1No license
  • +2Single-language instructions
  • +5Description quotes 18 example trigger phrases
  • +4Description says when NOT to use the skill
  • +4Structure: 48 headings
  • +3Step-by-step instructions: 105 items
  • +4Has examples (23 code blocks)
  • +4Reference files are cited in the instructions (2 of 6)

Quality base 70; lint remarks subtract, signals add up to 100. Result: 48.