Med Political

Political News Reporting

This skill specializes med-news-reporter for the political beat. Read med-news-reporter first for the general 6-step workflow (type selection, material audit, fact-check, balance, ethics, literacy). This file adds political-beat-specific discipline on top.

Overview

Distilled from political-journalism curricula at NCCU 政治系、NTU 政治系、Columbia Journalism School (Politics), Medill (Politics & Policy), Sciences Po École de journalisme. Covers the four main political-news sub-types: election / legislative / policy / official-statement reporting.

IRON LAW: No Stealth Bias

Political reporting carries stance through dozens of micro-choices that
each pass a literal fact-check but together push the reader toward a
conclusion: which adjective ("controversial" / "embattled" / "popular"),
whose quote leads, who gets the last word, where the contextualizing
data sits, which framing ("crackdown" vs "enforcement"), which timeline
("after the scandal" vs "during the second term"). Every such choice
must be either deliberately neutral or its slant disclosed. The reader
is owed a piece they can read without absorbing your verdict by osmosis.

Default LLM behavior is to pick the most "natural-sounding" word,
which in political contexts almost always carries the dominant-media
frame as bias. Override that default explicitly.

Why this is non-obvious: the literal-fact-check passes ("the bill failed 35-40, the polling shows 47%") while the cumulative effect of word-choice and ordering quietly endorses a side. This is the failure mode that gets political reporting accused of bias by every side — usually correctly.

Rationalization Table — these justifications DO NOT override the Iron Law:

When to Use

Trigger conditions:

• User supplies political material — 立法院質詢、政府記者會、政策白皮書、候選人政見、選戰文宣、民調數據、投票記錄、洩密文件、政治人物訪談 — and asks for a news piece.
• User asks for "選戰報導" / "政策分析" / "立委質詢整理" / "election coverage" / "policy tracker" / "political profile" / "campaign news".
• User paraphrases: "幫我把這份質詢做成新聞", "寫一篇選舉觀察", "整理成政策追蹤", "draft a piece on this vote", "analyze this poll".

Input signals:

• Named politicians, parties, government agencies, election candidates, civic organizations.
• Poll numbers, vote tallies, legislative bill numbers, policy documents.
• Direct quotes from officials, candidates, campaign staff, opposition figures.

When NOT to use:

• Government / agency press release in the agency's own voice → use pr-press-release.
• Op-ed / commentary / column on a political topic → use med-news-reporter with type=opinion (this skill is for news, not commentary; though Gotchas here still apply).
• Pure international relations / diplomacy without domestic-political angle → use med-international.
• Election campaign marketing / ads → not a journalism use case; decline.

Methodology

Step 0: Defer general workflow to med-news-reporter

Read or have already loaded med-news-reporter for: material audit, fact-checking, source-strength tagging, balance principle, media-ethics check, media-literacy self-check. Do not re-implement those steps here. This file specializes Steps 1, 2, 3, and adds a political-specific Step 7 (Frame Neutrality Audit).

Step 1: Classify the political-story sub-type

If the material spans sub-types (e.g. a candidate's policy speech), classify by the primary news driver, not the genre.

Step 2: Source vetting & stance tagging

Every named political actor in the piece must carry a stance tag at first mention, and the tag must be specific enough to let readers calibrate the quote:

• ❌ Vague: 「立委王大明表示...」
• ✅ Specific: 「民進黨立委、社會福利及衛生環境委員會召委王大明表示...」
• ✅ When stake matters: 「兼任 X 公司獨立董事的國民黨立委王大明表示...」(if the bill affects X)

Source tier tagging (extends med-news-reporter's tiering with political-specific tiers):

Step 3: Political-specific risk check

Beyond med-news-reporter's general ethics check, add:

1. 公職人員選舉罷免法 considerations:

   • Election period restrictions on poll publication (typically 投票日前 10 日禁止公布民調).
   • Equal-treatment principle for candidates: if you give Candidate A 300 words of accusation, Candidate B's response gets equivalent space, not a one-sentence dismissal.

2. 誹謗 (刑法 310) in political context:

   • Negative claims about politicians have heightened "public interest" defense, BUT the defense requires reasonable belief in truth at time of publication and public-benefit purpose. Do not assume "it's a politician, so I can say anything".
   • Allegations from opposition figures are still allegations; report them as accusations, not facts: 「對手陣營指稱」not 「事實上王大明...」.

3. 預算/財政數字 must include base-year, scope, and source:

   • ❌ "撥款 50 億" (over what period? from which budget? confirmed or proposed?)
   • ✅ "編列 113 年度公務預算 50 億元,屬經常門" / "三年期計畫累計 50 億"

4. 政府機關保密 vs 新聞自由:

   • 機密文件報導:確認文件真實性 + 報導具公益性 + 來源同意 + 給對方回應機會。
   • 偵查不公開:涉及檢調案件時,即使取得資訊,也需衡量是否造成偵查阻礙或當事人權益侵害。

Step 4: Frame Neutrality Audit (political-specific addition)

Before output, sweep the draft for:

1. Adjective audit: highlight every evaluative adjective ("controversial", "embattled", "popular", "hardline", "moderate", "extreme"). For each: replace with the specific underlying fact, OR cite who calls them that, OR delete.
2. Quote-order audit: count first-quote-from-each-side. If the same side leads in 3+ paragraphs, restructure.
3. Last-word audit: who gets the last quote? In a contested piece, the last word should be either neutral or alternated across pieces in a series.
4. Frame-word check: "crackdown" vs "enforcement", "regime" vs "government", "scheme" vs "program", "claim" vs "say". Each pair carries a stance. Use the more neutral term unless the loaded term is itself the news.
5. Implied causation: "after the scandal" implies the scandal caused the next event. Prefer "in date X" or explicit causal evidence.
6. Background-context placement: where you place the contextualizing fact (high vs low in the story) shapes the reader's frame. Place the most-relevant counter-context near the strongest claim.