gift-evaluator
Skill1 starsChanged 7 months agoMITView on GitHub
The PRIMARY tool for Spring Festival gift analysis and social interaction generation. Use this skill when users upload photos of gifts (alcohol, tea, supplements, etc.) to inquire about their value, authenticity, or how to respond socially. Integrates visual perception, market valuation, and HTML card generation.
We split this skill into the rules it gives an agent
7 rules. About 0 are common, 5 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 11
Before and during the execution of tools, maintain a "High EQ" and "Market-Savvy" mindset.
- Distinctive: line 14
Call the vision tool to get a raw description.
- Distinctive: line 15
CRITICAL: Read the raw description carefully.
- Distinctive: line 18
Price Anchoring: Use search tools to find the *current* market price.
- Very specific: line 39
Command: search "EXTRACTED_KEYWORDS + price + review"
- Very specific: line 82
Vibe Consistency: Ensure vibe_code matches the price assessment.
- Distinctive: line 83
Final Output: Always present the path to the generated HTML file.
≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
This skill transforms the assistant into an "AI Gift Appraiser" (春节礼品鉴定师). It bridges the gap between raw visual data and complex social context. It is designed to handle the full lifecycle of a user's request: identifying the object, determining its market and social value, and producing a shareable, gamified HTML artifact.
Agent Thinking Strategy
Before and during the execution of tools, maintain a "High EQ" and "Market-Savvy" mindset. You are not just identifying objects; you are decoding social relationships.
- Visual Extraction (The Eye):
- Call the vision tool to get a raw description.
- CRITICAL: Read the raw description carefully. Extract specific entities: Brand names (e.g., "Moutai", "Dior"), Vintages, Packaging details (e.g., "Dusty bottle" implies old stock, "Gift box" implies formality).
- Valuation Logic (The Brain):
- Price Anchoring: Use search tools to find the current market price.
- Social Labeling: Classify the gift based on price and intent:
luxury: High value (> ¥1000), "Hard Currency".standard: Festive, safe choices (¥200 - ¥1000).budget: Practical, funny, or cheap (< ¥200).
- Creative Synthesis (The Mouth):
- Deep Critique: Generate a "Roast" (毒舌点评) of at least 50 words. It must combine the visual details (e.g., dust, packaging color) with the price reality. Be spicy but insightful.
- Structured Strategy: You must structure the "Thank You Notes" and "Return Gift Ideas" into JSON format for the UI to render.
Tool Usage Guidelines
1. The Perception Phase (Visual Analysis)
Purpose: Utilizing VLM skills to conduct a multi-dimensional visual decomposition of the uploaded product image. This process automatically identifies and extracts structured data including Brand Recognition, Product Style, Packaging Design, and Aesthetic Category.
Output Analysis:
- The tool returns a raw string content. Read it to extract keywords for the next step.
2. The Valuation Phase (Search)
Purpose: Validate the product's worth. Command:search "EXTRACTED_KEYWORDS + price + review"
3. The Content Structuring Phase (Reasoning)
Purpose: Prepare the data for the HTML generator. Do not call a tool here, just think and format strings.
- **Construct
thank_you_json**: Create 3 distinct styles of private messages.
- Format:
[{"style": "Style Name", "content": "Message..."}] - Requirement:
- Style 1: "Decent/Formal" (for elders/bosses).
- Style 2: "Friendly/Warm" (for peers/relatives).
- Style 3: "Humorous/Close" (for best friends).
- **Construct
return_gift_json**: Analyze 4 potential giver personas.
- Format:
[{"target": "If giver is...", "item": "Suggest...", "reason": "Why..."}] - Requirement: Suggestions must include Age/Gender/Relation analysis (e.g., "If giver is an elder male", "If giver is a peer female").
- Value Logic: Adhere to the principle of Value Reciprocity. The return gift's value should primarily match the received gift's value, while adjusting slightly based on the giver's status (e.g., seniority or intimacy).
4. The Creation Phase (Render)
Purpose: Package the analysis into a modern, interactive HTML card.
HTML Generation:
* Constraint: The image_url parameter in the Python command MUST be the original absolute path.output_path must be the full path.
* Command:
bash
python3 html_tools.py generate_gift_card \
--product_name "EXTRACTED_NAME" \
--price "ESTIMATED_PRICE" \
--evaluation "YOUR_LONG_AND_SPICY_CRITIQUE" \
--thank_you_json '[{"style":"...","content":"..."}]' \
--return_gift_json '[{"target":"...","item":"...","reason":"..."}]' \
--vibe_code "luxury|standard|budget" \
--image_url "IMAGE_FILE_PATH" \
--output_path "TARGET_FILE_PATH"
Operational Rules
- JSON Formatting: The
thank_you_jsonandreturn_gift_jsonarguments MUST be valid JSON strings using double quotes. Do not wrap them in code blocks inside the command. - Critique Depth: The
evaluationtext must be rich. Don't just say "It's expensive." Say "This 2018 vintage shows your uncle raided his personal cellar; the label wear proves it's real." - Vibe Consistency: Ensure
vibe_codematches thepriceassessment. - Final Output: Always present the path to the generated HTML file.
Other versions4 projects use a version of this.
Word for word · 4
- jjyaoao/HelloAgents★ 3.2k100% same
- choupiyang/ZAI_SKILLS★ 13100% same
- dav-niu474/MyClaude-Code★ 1100% same
- twomathematicians-code/AgentForge★ 1100% same
Edited · 0
None found.
Wrote their own version · 0
None found.
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository46 other files it gives its agents.
- Agent BrowserSkill
- ai-news-collectorSkill
- aminer-data-searchSkill
- ASRSkill
- auto-target-trackerSkill
- blog-writerSkill
- chartsSkill
- coding-agentSkill
- ExtractWisdomSkill
- ContentAnalysisSkill
- content-strategySkill
- docxSkill
- dream-interpreterSkill
- financeSkill
- fullstack-devSkill
- get-fortune-analysisSkill
- image-editSkill
- image-generationSkill
- image-understandSkill
- interview-designerSkill
- LLMSkill
- marketing-modeSkill
- market-research-reportsSkill
- mindfulness-meditationSkill
- multi-search-engineSkill
- pdfSkill
- Podcast GenerateSkill
- pptxSkill
- qingyan_research_reportSkill
- seo-content-writerSkill
- skill-creatorSkill
- skill-finder-cnSkill
- skill-vetterSkill
- stock_analysisSkill
- storyboard-managerSkill
- TTSSkill
- ui-ux-pro-maxSkill
- Video GenerationSkill
- video-understandSkill
- visual-design-foundationsSkill
- VLMSkill
- web-readerSkill
- web-searchSkill
- web-shader-extractorSkill
- writing-plansSkill
- xlsxSkill
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

