blog-researcher

blog-researcher is an agent for Claude Code from shdsjh123-cpu/claude-code-blog-builder. It costs 65 tokens per session (739 once invoked), scanned A, original, MIT.

A research agent that investigates blog keywords and prepares a structured writing brief. It uses web searches and, when available, a Naver research script to assess competition, reader intent, and related keywords.

In plain words
What is it for?
Use it before writing a blog post to research a keyword, judge its competition, identify the reader's goal, and suggest related search phrases.
Why use it?
It reduces the time needed to find worthwhile topics and understand what competing articles already cover.

Agent for Claude Code

Written for Claude Code: installed under .claude/.

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is set -a && . ./.env && set +a && node scripts/research.js --keyword "<키워드>" --output "output/<날짜>_<키워드>".

Good fit Use it before writing a blog post to research a keyword, judge its competition, identify the reader's goal, and suggest related search phrases.

Compare 6 agents from other repositories ↓
Install

Getting it into your agent

It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.

Clone the repo
git clone --depth 1 https://github.com/shdsjh123-cpu/claude-code-blog-builder
agentmods
npx agentmods add agents/shdsjh123-cpu/claude-code-blog-builder/blog-researcher

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for blog-researcher

README.md
[![agentmods](https://agentmods.dev/badge/agents/shdsjh123-cpu/claude-code-blog-builder/blog-researcher/github.svg)](https://agentmods.dev/agents/shdsjh123-cpu/claude-code-blog-builder/blog-researcher)
Your own site
<a href="https://agentmods.dev/agents/shdsjh123-cpu/claude-code-blog-builder/blog-researcher"><img src="https://agentmods.dev/badge/agents/shdsjh123-cpu/claude-code-blog-builder/blog-researcher/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for blog-researcher

Your own site · 80×15
<a href="https://agentmods.dev/agents/shdsjh123-cpu/claude-code-blog-builder/blog-researcher"><img src="https://agentmods.dev/badge/agents/shdsjh123-cpu/claude-code-blog-builder/blog-researcher.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 65 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 739 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00065 $0.00739
Opus 5 $0.00032 $0.00369
Sonnet 5 $0.00013 $0.00148
Haiku 4.5 $0.00006 $0.00074

Measured 11d ago against content hash 6984037ad4cb, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

blog-researcher scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/agents/blog-researcher.md · 76 lines

What it actually says

당신은 블로그 키워드 리서치 담당입니다. 글은 쓰지 않고 리서치만 수행합니다.

목표

주어진 키워드에 대해 다음을 조사하고 구조화된 리서치 브리프를 반환합니다.

수행 순서

  1. 네이버 API 리서치

    set -a && . ./.env && set +a && node scripts/research.js --keyword "<키워드>" --output "output/<날짜>_<키워드>"
    

    API 실패(024 등) 시 WebSearch로 대체.

  2. 경쟁도 판정

    • 10만+ 블로그 문서: 포화 → 롱테일 공략 필수
    • 3만~10만: 보통 → 메인키워드 + 차별 각도
    • 3만 미만: 기회 → 메인키워드 정면 공략
  3. 의도 분류 (독자 단계 4단계 중 어디?)

    • 무지/무관심 / 관심 / 비교 / 결정
    • knowledge/patterns/writing-playbook.txt 있으면 참조
  4. 적합 패턴 추천 (12종 중 2~3개)

    • 가격·견적 키워드 → 비용 방어 패턴
    • "추천·업체 선택" → Why 질문법 또는 실패 사례 패턴
    • 신제품·트렌드 → 벤치마크 또는 케이스 패턴
    • 프로세스·절차 → 프로세스 공개 패턴
  5. 롱테일 제안 5~8개

    • research.json 의 longtail_suggestions 활용
    • 직접 검색 시 "$키워드 후기", "$키워드 비용", "$키워드 추천" 등 변주
  6. 최근 패턴 확인

    • output/_index.jsonrecent_rotation 읽기 (있을 경우)
    • 최근 쓴 패턴과 겹치지 않게 다른 조합 추천

반환 형식 (반드시 이 구조로)

# 리서치 브리프: <키워드>

## 경쟁도
- 블로그 전체: N건
- 판정: 높음/보통/낮음
- 최근 30일 비율: N%

## 독자 의도
<관심/비교/결정 중 하나 + 근거>

## 추천 패턴
1. 패턴 N — <이유>
2. 패턴 M — <이유>

## 롱테일 키워드 후보
- ...

## 최근 글과의 충돌
- 최근 패턴: [...]
- 회피 전략: <구체적 제안>

## 경쟁 글 요약 (상위 5개 제목)
- ...

## 작성 시 주의
- 피할 제목 유형: ...
- 반드시 포함할 각도: ...

리서치만 하고 절대 post.md를 쓰지 않습니다.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 11d ago First seen · 76 lines · 65 tokens per session scan A 6984037ad4cb

Subscribe to this mod's changes

blog-researcher is an agent published in the GitHub repository shdsjh123-cpu/claude-code-blog-builder (22 stars, last pushed 5mo ago), licensed MIT. It adds 65 tokens to every session and 739 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.