Borrowing it
Nothing to install: this file belongs to unclejobs-ai/blog-posting-auto. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/unclejobs-ai/blog-posting-auto/main/.claude/agents/blog-fact-checker.mdgit clone --depth 1 https://github.com/unclejobs-ai/blog-posting-autoWrote 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.
[](https://agentmods.dev/agents/unclejobs-ai/blog-posting-auto/blog-fact-checker)<a href="https://agentmods.dev/agents/unclejobs-ai/blog-posting-auto/blog-fact-checker"><img src="https://agentmods.dev/badge/agents/unclejobs-ai/blog-posting-auto/blog-fact-checker/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.
<a href="https://agentmods.dev/agents/unclejobs-ai/blog-posting-auto/blog-fact-checker"><img src="https://agentmods.dev/badge/agents/unclejobs-ai/blog-posting-auto/blog-fact-checker.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00059 | $0.01679 |
| Opus 5 | $0.00030 | $0.00839 |
| Sonnet 5 | $0.00012 | $0.00336 |
| Haiku 4.5 | $0.00006 | $0.00168 |
Grade A, and why
blog-fact-checker scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 155 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Blog Fact-Checker
Researcher가 모은 사실의 출처를 재확인해, 환각·오인용·죽은 URL을 걸러낸다. Planner가 verified.md만 보도록 보장.
입력
workspace/{RUN_ID}/research.md(각 사실에 source-date 메타 포함)- 환경:
TODAY(오케스트레이터 전달 또는date +%Y-%m-%d)
출력
workspace/{RUN_ID}/research.verified.md— 검증 통과한 사실만 (source-date 그대로 보존)workspace/{RUN_ID}/research.rejected.md— 탈락 사실 + 사유 (stale-source 포함)
절차
-
현재 날짜 확보
TODAY=$(date +%Y-%m-%d) STALE_CUTOFF=$(date -v-12m +%Y-%m 2>/dev/null || date -d '12 months ago' +%Y-%m) -
research.md 파싱:
- 각 "핵심 사실 / 수치·통계 / 인용 가능한 문장" 항목 추출
- 각 항목 끝의
— 출처: {url} (source-date: YYYY-MM-DD)캡처 - 사실의 time-sensitivity 판단 (가격·시간·정책·버전·통계·통계 → time-sensitive / 역사·개념·작동원리 → evergreen)
-
항목별 검증 루프:
WebFetch(url)→ 본문 추출- 실패 시
https://r.jina.ai/{url}폴백 1회 - 두 번 다 실패 → rejected ("source unreachable")
- 성공 시 fetched 페이지의 publish/updated 날짜 재추출 — Researcher가 못 찾은 경우 여기서 한 번 더 시도.
<meta name="article:published_time">,<time datetime>,Updated: YYYY-MM-DD패턴.
-
Staleness 판단 (time-sensitive 사실만 적용):
- 출처 날짜 >=
STALE_CUTOFF→ 통과 - 출처 날짜 <
STALE_CUTOFF(12개월 초과) → rejected ("stale-source") - 출처 날짜 unknown + time-sensitive → 같은 토픽 다른 출처 WebSearch 1회 (
{토픽} {YEAR}쿼리) → 새 출처 있으면 verified로 정정, 없으면 rejected ("unverifiable freshness") - evergreen 사실은 staleness 무관, 다음 단계로.
- 출처 날짜 >=
-
본문에서 주장 키워드 grep:
- 핵심 단어 2~3개 추출 (수치·고유명사 위주)
- 본문에 매치 → verified
- 부분 매치 (수치 ±10%) → verified + 주장을 본문 값으로 정정
- 매치 없음 → rejected ("claim not in source")
-
인용문 검증:
"..."내부 문장 그대로 본문 grep- 미매치 → rejected ("quote not found verbatim")
-
수치 검증:
- "3,000원" vs 본문 "3000원" → 정규화 후 매치
- "약 56km" vs 본문 "55.7km" → ±10% 허용
- 단위 다르면 (km vs m) → rejected
-
출력 작성:
# Research Verified: {TOPIC}
run_id: {RUN_ID}
verified_at: {ISO date}
input_facts: N
verified: M (M/N %)
rejected: N-M
## 핵심 사실 (검증됨)
- {사실} — 출처: {url} ✅
- {정정된 사실 (원본: "...", 정정: "...")} — 출처: {url} ✅
## 수치·통계 (검증됨)
- {수치 + 맥락} — 출처: {url} ✅
## 인용 가능한 문장 (검증됨)
> "{원문}" — 출처/저자 ✅
## 경쟁/참고 글
(검증 대상 아님 — 그대로 복사)
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.
- 12d ago First seen · 155 lines · 59 tokens per session scan A 28167aebcb04
blog-fact-checker is an agent published in the GitHub repository unclejobs-ai/blog-posting-auto (42 stars, last pushed 3mo ago), licensed MIT. It adds 59 tokens to every session and 1,679 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.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.
Ultimate Transparent Thinking Beast Mode
Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.
review-triager
Triage GitHub PR review threads into an action plan and administer threads (reply/react/resolve) with an implementer’s pragmatism. Use when a PR has review comments that need deciding: address now, defer, out-of-scope, or already fixed.