技能详情（站内镜像，无评论）
作者：ASP @anugotta
许可证：MIT-0
MIT-0 ·免费使用、修改和重新分发。无需归因。
版本：v1.0.0
统计：⭐ 0 · 25 · 0 current installs · 0 all-time installs
⭐ 0
安装量（当前） 0
🛡 VirusTotal ：良性 · OpenClaw ：良性
Package：anugotta/catalog-sku-matcher-india
安全扫描（ClawHub）
- VirusTotal ：良性
- OpenClaw ：良性
OpenClaw 评估
Instruction-only SKU-matching guidance whose requirements and instructions are coherent with its stated purpose; it requests no credentials, makes no installs, and stays within the problem domain.
目的
Name, description, and provided files (matching rules, scoring, examples, setup, validation) all align: the skill is focused on catalog SKU matching and normalization and does not request unrelated capabilities or resources.
说明范围
SKILL.md and the supplemental docs contain only matching strategies, rulebooks, scoring, setup and validation checklists. There are no instructions to read system files, access environment variables, call external endpoints, or transmit data outside the matching workflow.
安装机制
No install spec and no code files — this is instruction-only, so nothing is downloaded or executed on install and nothing is written to disk by the skill itself.
证书
No required environment variables, credentials, or config paths are declared. The requested resources are minimal and proportional to an instruction-only matching guide.
持久
Flags show always:false and default agent invocation behavior. The skill does not request persistent presence or modification of other skills or system-wide settings.
综合结论
This skill is a set of guidelines and rule files (no code or installs), so the immediate risk is low. Before using in production: (1) implement the rules in your own audited code rather than blindly executing external code, (2) build and run labeled validation sets as the setup and validation checklists recommend, (3) keep conservative auto-match thresholds and human review for ambiguous cases, (4) ensure any real data you feed into an impleme…
安装（复制给龙虾 AI）
将下方整段复制到龙虾中文库对话中，由龙虾按 SKILL.md 完成安装。
请把本段交给龙虾中文库（龙虾 AI）执行：为本机安装 OpenClaw 技能「Catalog SKU Matcher India」。简介：Match and normalize product listings across Indian ecommerce catalogs with vari…。
请 fetch 以下地址读取 SKILL.md 并按文档完成安装：https://raw.githubusercontent.com/openclaw/skills/refs/heads/main/skills/anugotta/catalog-sku-matcher-india/SKILL.md
（来源：yingzhi8.cn 技能库）SKILL.md
---
name: catalog-sku-matcher-india
description: Match and normalize product listings across Indian ecommerce catalogs with variant-aware rules, confidence scoring, false-match prevention, and review queues for ambiguous pairs.
metadata: {"openclaw":{"emoji":"🧩","homepage":"https://clawhub.ai/Michael-laffin/price-tracker"}}
---
# Catalog SKU Matcher India
## Purpose
Build reliable cross-store product matching for Indian catalogs so price comparison is accurate.
## Disclaimer
This skill provides matching and normalization guidance only. It does not guarantee perfect match accuracy for all catalogs or seller data quality.
Use at your own risk. The skill author/publisher/developer is not liable for direct or indirect loss, incorrect match decisions, trading losses, or other damages arising from use or misuse of this guidance.
## Matching strategy
Use a layered approach:
1. **Hard identifiers**
- model number / GTIN / MPN / ISBN where available
2. **Variant normalization**
- brand
- model family
- storage/RAM
- color
- size/pack quantity
- condition (new/refurbished/used)
3. **Soft similarity**
- token similarity on cleaned title
- key-attribute overlap
- seller metadata sanity checks
4. **Confidence score**
- `high`: auto-match
- `medium`: human review queue
- `low`: reject
## False-match guardrails
- Never match different storage/RAM variants as same SKU.
- Never match bundles/accessories to standalone products.
- Never ignore refurbished/used condition differences.
- Require manual review when two or more variant fields are missing.
## Output format
When matching listings, return:
1. canonical SKU candidate
2. matched listings with confidence level
3. rejected candidates with reason codes
4. manual review queue entries
## Setup
Read [setup.md](setup.md) and define normalization dictionaries first.
## Validation
Run [validation-checklist.md](validation-checklist.md) on labeled test sets before production.
## References
- Rules and reason codes: [matching-rules.md](matching-rules.md)
- Confidence scoring: [scoring-guide.md](scoring-guide.md)
- Edge-case examples: [examples.md](examples.md)