Produces structured neutral summaries of public debates, mapping positions, actors, evidence, and unresolved questions for journalists to quickly understand and find underreported angles.
How this skill is triggered — by the user, by Claude, or both
Slash command
/autopunk-media-skills:debate-state-summaryThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Produces a structured, neutral summary of the current state of a public debate or controversy — mapping the main positions, the actors behind them, the evidence each side cites, and the unresolved questions — so a journalist can get up to speed quickly and identify the underreported angles.
Produces a structured, neutral summary of the current state of a public debate or controversy — mapping the main positions, the actors behind them, the evidence each side cites, and the unresolved questions — so a journalist can get up to speed quickly and identify the underreported angles.
Required: The topic or question at the centre of the debate (be specific — "Should gig-economy workers be classified as employees?" is better than "gig economy debate"); the jurisdiction or context if relevant.
Optional: Your current understanding of the main positions; specific actors or organisations you know are involved; the media narrative you have seen so far (helps the assistant identify what is being underreported); the audience for your output (general public, specialist readership, editorial team).
Defines the core question. Distils the debate to a single, precise question that the competing positions are answering differently. This prevents the summary from drifting into adjacent issues.
Maps the main positions. Identifies 3–5 distinct positions in the debate (not just "for" and "against" — most real debates have more nuance). For each position, describes: the argument in 2–3 sentences, the types of actors who hold it, and the evidence or reasoning they cite.
Identifies the points of agreement. Notes where opposing sides actually agree — these are often overlooked in coverage and can be the most revealing elements of a debate.
Flags the unresolved questions. Lists the specific factual or analytical questions that remain genuinely contested — where the evidence is ambiguous, missing, or disputed. These are the questions a journalist should investigate rather than assume answers to.
Highlights the underreported angles. Based on the positions mapped, identifies 2–3 aspects of the debate that receive less media attention than they deserve — minority positions, overlooked stakeholders, or evidence that complicates the dominant narrative.
600–900 words. Four sections: Core Question (one sentence), Positions Map (3–5 positions, each with argument, actors, and evidence), Points of Agreement (2–3 bullet points), and Unresolved Questions and Underreported Angles (combined section, 4–6 bullet points). Tone: strictly neutral — the summary should not reveal which position the journalist or the assistant favours. Language: accessible to a non-specialist reader, with technical terms defined on first use.
Topic: Whether national governments should mandate algorithmic transparency for social media platforms — specifically, requiring platforms to publish the criteria their recommendation algorithms use to rank and promote content Jurisdiction: Primarily a Western democratic context, but the debate is global Media narrative seen so far: Most coverage frames it as "regulators vs. big tech" — regulators want transparency, platforms resist
Core Question
Should governments require social media platforms to publicly disclose the specific criteria and weighting factors their recommendation algorithms use to rank, filter, and promote content to users?
Positions Map
Position 1 — Full algorithmic disclosure Platforms should be legally required to publish the complete logic of their recommendation algorithms, including the specific signals used (engagement metrics, user behaviour, advertiser preferences) and how they are weighted. Proponents argue that content recommendation is a form of editorial decision-making with enormous public impact, and that editorial decisions at this scale require public accountability. This position is held by many digital rights organisations, some academics in media studies and computer science, and several legislative proposals in democratic countries. Evidence cited: research showing that recommendation algorithms amplify divisive content, increase political polarisation, and can promote health misinformation at scale.
Position 2 — Structured transparency to regulators, not the public Platforms should be required to disclose algorithmic criteria to an independent regulator or auditor, but not to the general public. This position argues that full public disclosure would expose trade secrets, enable manipulation by bad actors (who could game the algorithm), and provide little practical benefit to ordinary users who cannot interpret technical specifications. Held by some technology policy researchers, certain industry-aligned think tanks, and some regulators who prefer audit-based models. Evidence cited: precedent in financial regulation, where banks disclose risk models to regulators rather than the public.
Position 3 — Outcome-based regulation, not process disclosure Instead of mandating transparency about how algorithms work, governments should set measurable standards for what outcomes algorithms must produce (e.g., limits on the amplification of verified misinformation, minimum exposure to diverse viewpoints) and hold platforms accountable for results. Proponents argue that algorithmic systems are too complex and fast-changing for static disclosure to be meaningful, and that outcomes are what actually matter to users. Held by some computer scientists, AI policy researchers, and platform companies that prefer performance standards to structural disclosure. Evidence cited: the rapidly evolving nature of machine-learning systems that makes any snapshot of algorithm logic obsolete within months.
Position 4 — No government mandate Algorithmic transparency should be voluntary, driven by market competition and user choice rather than regulation. Government mandates risk stifling innovation, creating compliance costs that benefit large incumbents over smaller competitors, and setting precedents for government control over information systems. Held by platform industry trade associations, some libertarian-leaning think tanks, and certain technology executives. Evidence cited: the history of regulatory capture in other industries, the difficulty of defining "algorithm" in a legally precise way, and the risk that transparency mandates in democratic countries legitimise censorship mandates in authoritarian ones.
Points of Agreement
Unresolved Questions and Underreported Angles
npx claudepluginhub ur-grue/autopunk-media-skills --plugin autopunk-media-skillsProduces a structured, neutral summary of a public debate or controversy, mapping positions, evidence, and crux points. Useful for journalists getting up to speed on contested topics.
Adversarial multi-round reasoning with blind-judge panel to reach rigorous conclusions. TRIGGER when: user wants rigorous reasoning or argument evaluation; user wants a decision analyzed from multiple angles; user wants devil's advocate critique; user asks "what are the strongest arguments for/against"; user wants a structured debate; user wants to avoid groupthink or anchoring; user invokes /autoresearch:reason. DO NOT TRIGGER when: user wants a simple recommendation; user wants a quick summary; user wants factual lookup; user just wants pros/cons without adversarial pressure.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.