Agent skill

Sentiment Lexicon Builder

by revfactory in revfactory/harness-100

감성 사전 구축, ABSA(Aspect-Based Sentiment Analysis) 설계, 감성 점수 보정, 도메인 특화 감성 분석 방법론 가이드.

Apache-2.0Auto-check passedSales & Support

Install Sentiment Lexicon Builder

skills CLI
$ npx skills add revfactory/harness-100 --skill sentiment-lexicon-builder -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install revfactory/harness-100 sentiment-lexicon-builder --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/revfactory/harness-100.git skills-src && mkdir -p .claude/skills && cp -r skills-src/ko/33-text-processor/.claude/skills/sentiment-lexicon-builder .claude/skills/sentiment-lexicon-builder && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
sentiment-lexicon-builder
GitHub stars
1.3k
Token cost
~1.1k tokens
SKILL.md length
108 words
Files
1
Skills in repo
464
Repo updated
First seen
Licence
Apache-2.0

At a glance

감성 사전 구축, ABSA(Aspect-Based Sentiment Analysis) 설계, 감성 점수 보정, 도메인 특화 감성 분석 방법론 가이드.

  • Tasks that involve Customer feedback analysis
  • SKILL.md covers 감성 분석 접근법 비교, 감성 사전 구축, ABSA (Aspect-Based Sentiment… and 감성 점수 보정, plus 2 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Sentiment Lexicon Builder is an agent skill from revfactory/harness-100. 감성 사전 구축, ABSA(Aspect-Based Sentiment Analysis) 설계, 감성 점수 보정, 도메인 특화 감성 분석 방법론 가이드. '감성 사전', '감성 분석 모델', 'ABSA', '측면별 감성', '감성 점수', '극성 사전', '도메인 감성', '감정 분류' 등 감성분석 설계 시 이 스킬을 사용한다. sentiment-analyzer의 감성분석 역량을 강화한다. 단, 텍스트 전처리나 보고서 작성은 이 스킬의 범위가 아니다.

Its SKILL.md is about 1.1k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Sales & Support, covering Customer feedback analysis. The licence is Apache-2.0.

When your agent uses it

  • Tasks that involve Customer feedback analysis

Example prompts

  • “/sentiment-lexicon-builder”

Requirements

  • Python 3

What it can do on your machine

Read from SKILL.md and the folder at commit 8e8d35c. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md (its code samples are python, yaml and markdown).

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Sentiment Lexicon Builder loads about 1.1k tokens when it runs. Until then it costs about 70 tokens; SKILL.md has 108 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~70
When it runs · the whole SKILL.md, loaded when a task matches
~1.1k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from revfactory/harness-100 at commit 8e8d35c, republished under its Apache-2.0 licence (© revfactory). 108 words, ~1,088 tokens.

Download SKILL.mdSave it as .claude/skills/sentiment-lexicon-builder/SKILL.md (or your agent's skills folder).
name
sentiment-lexicon-builder
description
감성 사전 구축, ABSA(Aspect-Based Sentiment Analysis) 설계, 감성 점수 보정, 도메인 특화 감성 분석 방법론 가이드. '감성 사전', '감성 분석 모델', 'ABSA', '측면별 감성', '감성 점수', '극성 사전', '도메인 감성', '감정 분류' 등 감성분석 설계 시 이 스킬을 사용한다. sentiment-analyzer의 감성분석 역량을 강화한다. 단, 텍스트 전처리나 보고서 작성은 이 스킬의 범위가 아니다.

Sentiment Lexicon Builder — 감성 사전 및 ABSA 설계 가이드

도메인 특화 감성 분석 시스템을 설계하고 구축하는 방법론.

감성 분석 접근법 비교

접근법장점단점적합
사전 기반빠름, 해석 가능도메인 한계, 문맥 무시소규모, 빠른 프로토타입
ML 기반 (전통)도메인 적응학습 데이터 필요레이블 데이터 있을 때
딥러닝 (BERT)문맥 이해, 높은 정확도리소스 필요대규모, 정확도 중시
LLM (프롬프트)제로샷, 유연비용, 속도다양한 도메인, 소량

감성 사전 구축

한국어 기본 사전
python
SENTIMENT_LEXICON = {
    # 긍정 (1.0 ~ 0.1)
    "좋다": 0.8, "훌륭하다": 0.9, "만족": 0.7, "추천": 0.8,
    "편리하다": 0.7, "깔끔하다": 0.6, "최고": 0.9, "친절하다": 0.8,
    "빠르다": 0.6, "저렴하다": 0.5,

    # 부정 (-0.1 ~ -1.0)
    "나쁘다": -0.8, "불만": -0.7, "실망": -0.8, "느리다": -0.6,
    "비싸다": -0.5, "불편하다": -0.7, "최악": -0.9, "불친절": -0.8,
    "고장": -0.7, "환불": -0.6,

    # 강도 수정자
    "매우": 1.5,    # 강화
    "약간": 0.5,    # 약화
    "정말": 1.5,
    "조금": 0.5,
    "너무": 1.3,    # 문맥 따라 긍/부정 모두
}

NEGATION_WORDS = {"않다", "않", "못", "없다", "없", "안"}
도메인 특화 사전 자동 구축
python
def build_domain_lexicon(corpus, labels, base_lexicon, top_n=200):
    """
    TF-IDF + PMI 기반 도메인 감성 사전 자동 구축

    1. 긍정/부정 리뷰에서 각각 TF-IDF 상위 단어 추출
    2. PMI(Pointwise Mutual Information)로 감성 극성 계산
    3. 기존 사전과 병합
    """
    pos_texts = [t for t, l in zip(corpus, labels) if l == 'positive']
    neg_texts = [t for t, l in zip(corpus, labels) if l == 'negative']

    # 각 클래스에서의 출현 확률
    for word in vocabulary:
        p_word = count(word, corpus) / len(corpus)
        p_pos = count(word, pos_texts) / len(pos_texts)
        p_neg = count(word, neg_texts) / len(neg_texts)

        pmi_pos = log2(p_pos / p_word) if p_pos > 0 else 0
        pmi_neg = log2(p_neg / p_word) if p_neg > 0 else 0

        polarity = pmi_pos - pmi_neg  # 양수=긍정, 음수=부정

    return domain_lexicon

ABSA (Aspect-Based Sentiment Analysis)

설계 구조
입력: "배송은 빠른데 제품 품질이 별로예요"

1. 측면(Aspect) 추출:
   - "배송" → [배송/서비스]
   - "품질" → [제품/품질]

2. 측면별 감성 분석:
   - 배송: "빠른" → 긍정 (0.6)
   - 품질: "별로" → 부정 (-0.7)

3. 결과:
   {
     "overall": -0.05,
     "aspects": {
       "배송": {"sentiment": "positive", "score": 0.6, "keywords": ["빠른"]},
       "품질": {"sentiment": "negative", "score": -0.7, "keywords": ["별로"]}
     }
   }
측면 카테고리 설계 (이커머스 예시)
yaml
aspects:
  제품:
    품질: [품질, 퀄리티, 소재, 재질, 마감, 내구성]
    디자인: [디자인, 색상, 색깔, 모양, 외관]
    사이즈: [사이즈, 크기, 사이즈감, 피팅]
    가격: [가격, 가성비, 비싸, 저렴, 합리적]
  서비스:
    배송: [배송, 택배, 도착, 배달]
    포장: [포장, 박스, 패키지]
    교환환불: [교환, 환불, 반품, AS]
    고객응대: [응대, 상담, 친절, 불친절]

감성 점수 보정

부정어 처리
python
def handle_negation(tokens, scores):
    """부정어 뒤 3토큰 이내 감성 반전"""
    negation_window = 0
    adjusted = []
    for token, score in zip(tokens, scores):
        if token in NEGATION_WORDS:
            negation_window = 3
        elif negation_window > 0:
            score = -score * 0.8  # 완전 반전이 아닌 80% 반전
            negation_window -= 1
        adjusted.append(score)
    return adjusted
강도 수정자 처리
python
def apply_intensifiers(tokens, scores):
    """강도 수정자에 따른 점수 조정"""
    adjusted = []
    for i, (token, score) in enumerate(zip(tokens, scores)):
        if i > 0 and tokens[i-1] in INTENSIFIERS:
            score *= INTENSIFIERS[tokens[i-1]]
        adjusted.append(score)
    return adjusted
이모지 감성 매핑
python
EMOJI_SENTIMENT = {
    "😊": 0.8, "😍": 0.9, "👍": 0.7, "❤️": 0.8, "🙏": 0.5,
    "😡": -0.9, "😤": -0.7, "👎": -0.8, "😢": -0.6, "💔": -0.7,
    "😐": 0.0, "🤔": -0.1,
}

감성 분석 평가 메트릭

python
# 감성 분류 평가
from sklearn.metrics import classification_report

print(classification_report(y_true, y_pred,
    target_names=['부정', '중립', '긍정']))

# ABSA 평가
# - 측면 추출: Precision, Recall, F1
# - 측면별 감성: Accuracy, Macro-F1
# - 전체: Micro-F1 (측면추출 + 감성 모두 정확해야 정답)

보고서 구조

markdown
## 감성 분석 결과

### 전체 요약
| 극성 | 건수 | 비율 |
|------|------|------|
| 긍정 | 650 | 65% |
| 중립 | 150 | 15% |
| 부정 | 200 | 20% |

### 측면별 감성
| 측면 | 긍정 | 부정 | 점수 | 주요 키워드 |
|------|------|------|------|-----------|
| 배송 | 80% | 10% | +0.6 | 빠른, 정확 |
| 품질 | 40% | 45% | -0.2 | 별로, 약함 |

### 시계열 추이
### 주요 부정 패턴 (액션 아이템)

© revfactory, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in ko/33-text-processor/.claude/skills/sentiment-lexicon-builder of revfactory/harness-100.

Open the folder on GitHubat commit 8e8d35c

Compare with similar skills

Sentiment Lexicon Builder next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Sentiment Lexicon Builder compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Sentiment Lexicon Builder this skillrevfactory/harness-1001.3k—~1.1kAutomated safety check: PassApache-2.0
Review Analysisliangdabiao/amazon-sorftime-research-MCP-skill9531 repos~2.5kAutomated safety check: PassNone
Bggg Data Amazonbinggandata/bggg-skills604—~1.4kAutomated safety check: PassMIT
Zsxqunnoo/zsxq-skill304—~3.8kAutomated safety check: PassMIT
Roadtrip NavigatorWaybox-AI/roadtrip-skill126—~3.4kAutomated safety check: PassMIT
Always Compareai-analyst-lab/ai-analyst304—~1.4kAutomated safety check: PassMIT

Similar skills

  • Review Analysis

    liangdabiao/amazon-sorftime-research-MCP-skill

    对亚马逊商品评论进行深度分析,自动识别产品痛点、分析退货原因,生成改进建议和客服回复模板。Invoke when user uses /review-analysis command with a product ASIN.

    953 GitHub starsUsed in 1 repo~2.5k tokens
    Sales & SupportAuto-check passed
  • Bggg Data Amazon

    binggandata/bggg-skills

    Collect Amazon.com written product reviews at scale through Woot's public review AJAX route, retain every attempt and error log, reconcile partial runs, and normalize exact review text into…

    604 GitHub stars~1.4k tokensUpdated 1 mo ago
    Sales & SupportAuto-check passed
  • Zsxq

    unnoo/zsxq-skill

    知识星球 CLI(zsxq-cli)与底层接口完整操作指南,涵盖星球和内容管理、Skill Pay 微信支付场景。当用户提到知识星球、zsxq、小密圈、星球、登录/认证、发帖、评论、回答、编辑、删除主题、定时发布/定时任务/定时回答、投票、问答主题、markdown 正文、AI…

    304 GitHub stars~3.8k tokensUpdated 18 days ago
    Sales & SupportAuto-check passed
  • Roadtrip Navigator

    Waybox-AI/roadtrip-skill

    Generate North American road-trip itineraries as a map-first, offline-friendly single-file HTML page.

    126 GitHub stars~3.4k tokensUpdated 1 mo ago
    Sales & SupportAuto-check passed
  • Always Compare

    ai-analyst-lab/ai-analyst

    Never present a metric or number in isolation; anchor every number to a comparison (prior period, benchmark, or another segment) or state that none is available.

    304 GitHub stars~1.4k tokensUpdated 9 days ago
    Sales & SupportAuto-check passed
  • Account Deletion

    gustavscirulis/snapgrid

    Generates an Apple-compliant account deletion flow with multi-step confirmation UI, optional data export, configurable grace period, Keychain cleanup, and server-side deletion request.

    117 GitHub starsUsed in 1 repo~2.5k tokens
    Sales & SupportAuto-check: notes

More from revfactory/harness-100

All 464 skills in this repo
  • Anti Bot Analyzer

    revfactory/harness-100

    A skill for analyzing website anti-bot defense mechanisms and developing legitimate evasion strategies.

    1.3k GitHub stars~1.1k tokensUpdated 6 mo ago
    Auto-check passed
  • API Error Design Patterns

    revfactory/harness-100

    Reference for designing how an API reports failures: structured error codes, response shapes, client-friendly messages, an error catalog and retry or fallback advice.

    1.3k GitHub stars~1.6k tokensUpdated 6 mo ago
    Auto-check passed
  • API Security Checklist

    revfactory/harness-100

    Walks a backend-dev agent through OWASP API Top 10 checks, authentication and authorization patterns, and defense code during API design.

    1.3k GitHub stars~1.7k tokensUpdated 6 mo ago
    Auto-check passed
  • Arg Parser Generator

    revfactory/harness-100

    Methodology for systematically designing and generating CLI tool argument parser structures.

    1.3k GitHub stars~1.2k tokensUpdated 6 mo ago
    Auto-check passed
  • Audience Segmentation

    revfactory/harness-100

    Audience segmentation skill used by the analyst and curator agents.

    1.3k GitHub stars~1.3k tokensUpdated 6 mo ago
    Auto-check passed
  • Audio Storytelling

    revfactory/harness-100

    Audio storytelling skill used by the podcast scriptwriter and show note editor.

    1.3k GitHub stars~1.6k tokensUpdated 6 mo ago
    Auto-check passed

Categories

Questions about Sentiment Lexicon Builder

What does Sentiment Lexicon Builder do?

감성 사전 구축, ABSA(Aspect-Based Sentiment Analysis) 설계, 감성 점수 보정, 도메인 특화 감성 분석 방법론 가이드. Sentiment Lexicon Builder is an agent skill from revfactory/harness-100. 감성 사전 구축, ABSA(Aspect-Based Sentiment Analysis) 설계, 감성 점수 보정, 도메인 특화 감성 분석 방법론 가이드.

When should I use Sentiment Lexicon Builder?

Sentiment Lexicon Builder fits situations like: tasks that involve Customer feedback analysis.

How do I install Sentiment Lexicon Builder in Claude Code?

Run `npx skills add revfactory/harness-100 --skill sentiment-lexicon-builder -a claude-code`. Or copy the skill folder (ko/33-text-processor/.claude/skills/sentiment-lexicon-builder in revfactory/harness-100) into .claude/skills/sentiment-lexicon-builder in your project. Claude Code loads it when a task matches its description.

How do I install Sentiment Lexicon Builder in Codex?

Run `npx skills add revfactory/harness-100 --skill sentiment-lexicon-builder -a codex`. Or copy the skill folder (ko/33-text-processor/.claude/skills/sentiment-lexicon-builder in revfactory/harness-100) into .agents/skills/sentiment-lexicon-builder in your project. Codex loads it when a task matches its description.

Can I use Sentiment Lexicon Builder in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add revfactory/harness-100 --skill sentiment-lexicon-builder -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/sentiment-lexicon-builder, .gemini/skills/sentiment-lexicon-builder, .github/skills/sentiment-lexicon-builder and .opencode/skills/sentiment-lexicon-builder in your project.

What does Sentiment Lexicon Builder need to run?

SKILL.md names no scripts, command-line tools or credentials: Sentiment Lexicon Builder is instructions for the agent only. Our summary lists: Python 3.

Does Sentiment Lexicon Builder access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Sentiment Lexicon Builder safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Sentiment Lexicon Builder use?

Sentiment Lexicon Builder is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Sentiment Lexicon Builder use?

About 1.1k tokens (SKILL.md is roughly 4.4k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Sentiment Lexicon Builder?

Skills that share tags, products or a category with Sentiment Lexicon Builder: Review Analysis (liangdabiao/amazon-sorftime-research-MCP-skill, 953 stars), Bggg Data Amazon (binggandata/bggg-skills, 604 stars), Zsxq (unnoo/zsxq-skill, 304 stars) and Roadtrip Navigator (Waybox-AI/roadtrip-skill, 126 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Sentiment Lexicon Builder?

revfactory (a GitHub user) maintains it in revfactory/harness-100, which has 1,295 GitHub stars. The repository holds 464 skills in this directory. The repository was last updated on March 22, 2026.

Source: revfactory/harness-100 on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.