> For the complete documentation index, see [llms.txt](https://thecontentforge.gitbook.io/thecontentforge-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://thecontentforge.gitbook.io/thecontentforge-docs/feature-glossary/glossary/trends-and-media.md). # Trends and Media Intelligence *Last updated: 2026-08-07.* This is how the product answers "what is happening right now, and what should we post about it". It watches trends, news and social media, scores what is worth your attention, and hands you angles, hooks and assets you can take straight into content creation. ## ForgeSays ForgeSays is the recurring briefing on what your audience is talking about, organised by the topics you choose to track. * **ForgeSays** - Runs a fresh intelligence cycle roughly every twelve hours for each of your tracked categories, blending live social trends with news coverage from your sources. Each run produces a short written summary plus three worked-up result sets, and older briefings drop off after a day so you are always looking at what is current. * **Viral Topics** - The ranked list of what is moving right now, each with an urgency score out of ten, why it matters, an angle written specifically for your brand, and a draft hook you can post. Topics also carry sentiment, relevant tickers, best platforms, the post format they suit such as a hot take or an explainer, and a breaking or rising label so you can tell a real spike from background noise. * **Post Angles** - Concrete post ideas rather than topics, each tagged with the platform to use, when to post it, expected engagement, and the reason that angle was chosen. Every angle comes with a ready prompt so you can go straight to drafting. * **Engagement Targets** - Specific accounts worth replying to, quoting or retweeting while a story is hot, with the reason each one matters and a suggested reply you can copy and adapt. This is the "join the conversation" half of the briefing rather than the "publish your own post" half. * **Verified** - A separate fact-check pass over the claims in the briefing, marking each one verified, partial, contested or unverified and listing the sources behind it. Claims that pass show a badge on the matching topic, claims it could not substantiate are called out in their own list, and you can re-run the check on demand, which matters when you are about to publish something time sensitive. * **Rewrites for other formats** - The same verification pass also returns the briefing rewritten as a newsletter paragraph and as an investor brief, so a social-first briefing can be reused for longer-form channels without starting again. * **Category management** - You decide what ForgeSays researches. Add up to twelve categories, give each one must-cover points and optional topics that steer the AI, add keywords that pull matching news into that category, reorder them, turn them on or off, and pick an icon and colour. Category names are translated automatically for multi-language workspaces, and only editors and above can change any of this. * **Run all, or run one** - Refresh every category at once or top up a single one, with progress showing per category so you are never blocked waiting on the whole set. * **Straight into drafting** - Every topic, angle and hook has copy, publish and send-to-Content-Forge controls, so a briefing item becomes a draft or a scheduled post without retyping anything. * **Briefing provenance** - Each briefing shows the time window it covers, how long it stays fresh, and how many items came from social trends, from news, or from both sources agreeing. It is an at-a-glance confidence read before you build content on top of it. ## ForgeSays for Web3 projects Web3 project workspaces get a different shape of ForgeSays: instead of broad topic briefings, it watches the specific accounts and feeds that matter to your project and tells you which posts are worth reacting to. * **Project opportunity radar** - A scored feed of things published by the sources you track, such as the chains you build on, ecosystem accounts, partners, competitors, founders, researchers and regulators. Each item shows a relevance score, a plain explanation of why it is an opportunity, and the entities or topics it matched, so you can act on the handful that matter instead of reading everything. * **Rescan sources and triage** - Trigger a fresh scan at any time, filter by source type, action, time window and minimum score, then generate a post from an item, queue it for later, or dismiss it. Dismissing and queuing keep the feed clean so the same item does not resurface. * **Action buckets** - The buckets opportunities get sorted into are yours to define, up to twelve of them, reorderable and switchable on or off. Set which source types and keywords land in each bucket and which kind of content it opens as when you send it to drafting, so a partner announcement and a competitor launch get treated differently by default. ## 12-Hour Trend Scan Available to every workspace, this is the wider "what is trending publicly" view, with the emphasis on bridging a trend to your brand rather than commenting on trends for their own sake. * **12-Hour Trend Scan** - The top twenty trends refreshed twice a day, each with a plain-language summary, an estimate of how many people are posting about it, and whether it is breaking, rising, steady or cooling. It shows when the last scan ran and when the next one is due, and if a scan fails you keep the previous results rather than an empty page. * **Brand fit scoring** - Every trend is rated for how well your brand can credibly join it: strong, moderate, low, or not recommended. You can open any rating to read the reasoning behind it, so a verdict is never a black box. * **Adapted angles** - For each trend you get angles written for your brand, each with the hook to open with, why it fits your brand, how the trend bridges to your subject, the impression opportunity, and a call to action. One click sends the chosen angle to drafting with the trend context and cautions attached. * **Personalize for my brand** - Until your brand voice is set up you see generic angles for each trend. Once it is set up, you can adapt the whole scan, or a single trend, to your voice, and re-personalize later if a trend moves on. * **Trend background** - Expand a trend to see why it is trending, what people are actually saying, related hashtags, the key accounts driving it, top posts, the regions it is running in, and the sources behind the summary. This is the context you need before deciding whether to touch a story. * **Risk flags** - Trends carry a risk level, and where relevant a written note about what makes them sensitive plus an explicit list of things to avoid saying. Those cautions travel with the angle into drafting. ## Market Signals Market Signals is the always-on news feed, grouped by story rather than by article, so ten write-ups of the same event arrive as one entry. * **Market Signals** - A live feed of clustered market stories. Each card shows the headline, a short AI summary, how many distinct sources are reporting it, the total number of mentions, the chains and sectors involved, and how recently it was seen. It is useful straight out of the box, before you configure anything. * **Rising** - A marker on stories that are genuinely developing, so you can tell a real story gathering pace from one outlet writing about itself. * **Top Signals, From My Sources and Saved** - Three views of the same feed. Sources your workspace has added are boosted and badged so your own sources visibly sharpen the results, and anything you save is kept in its own list. * **Chain filter and trending narratives** - Narrow the feed to a single chain, and see a strip of the narratives carrying the most momentum across everything currently in view. * **Per-story actions** - Expand a story to read every individual mention and jump to the original articles, save it, hide it so it stops appearing, or send the headline and summary into drafting. ## Media Radar Media Radar finds the images, videos, charts and article visuals that are performing across the sources you monitor, explains why each one is worth your attention, and gives you safe ways to build on it. It is available per workspace, so it appears only in workspaces that include it. * **Feed** - A ranked grid of trending media across your monitored sources, with save and dismiss controls on every card. Each card shows a thumbnail of the item, the platform and author, a recommendation, freshness and acceleration, fact-check status, a rights indicator, and score bars for overall quality, virality and relevance, plus a personalised ranking once the system has learned your preferences. Cards flag when the same item is being posted by many accounts or is part of a bigger running story, and say plainly when an item scored too low to build on and is there for research only. * **Feed filters** - Narrow by media type, by source, by time window from the last six hours out to seven days or any time, and by fact-check status. * **Angles others have not taken** - When you select a category, Media Radar surfaces the differentiated angles the crowd is not covering, marked as whitespace, emerging or saturated, each with an opportunity score, an explanation of why it is still open, and a suggested hook you can copy. * **Saved** - Your shortlist. Anything you save from the feed lands here, along with the branded cards and charts you generate. * **Categories** - Your monitors, listed with how often each one scans and whether it is running. Pause, resume, edit, duplicate or delete any of them, and mark a category as private to you or shared with your team. * **Category builder** - A full setup screen for a new monitor, where you give it a name and a context brief that steers the AI, set the keywords, exclusions, tickers, people and projects that route content into it, choose the media types and platforms, and add the specific accounts, searches, feeds, news queries and channels to watch. You also control fact-check strictness, brand safety level, minimum quality and virality thresholds, how recent an item must be, standing instructions for the research agent, scan frequency, and an alert threshold for standout finds. New monitors start filling the feed shortly after you save. * **Auto-discovered categories** - When a theme keeps trending across your sources and none of your monitors cover it, Media Radar proposes and adds a category for it, so a new narrative does not go unwatched just because nobody thought to set it up. * **Detail view** - Opening an item gives you its thumbnail, the caption, when it was first seen and published, who posted it, and the reasons it surfaced. Separate views show the full score breakdown including credibility, media quality, rights risk and fact confidence, and the wider story with the related items other sources published. * **Deep dive** - Ask the research agent to investigate a single item. It researches the topic across your feed, checks the facts, and comes back with a summary, recommendations for how to use it, and a list of what it flagged and why. It shows its working, and tells you plainly when it could only complete a partial scan. * **Make card** - Recreate the content of a post or article as a branded quote or headline card. It rebuilds the content in your own brand rather than screenshotting someone else's interface, and the finished card appears in your saved items. * **Make chart** - Recreate a market story as a branded, data-stamped chart. Pick one of the supported major crypto assets, the chart style, the time range and the output size, and the finished image lands in your saved items ready to post. * **More like this, less like this** - Thumbs up and down in an item's detail view teach your feed what your team actually wants. Personal taste shapes what you see first, and it can never disguise a weak source. * **Media Radar settings** - Choose the one or two channels you publish to, in priority order, which sets the aspect ratio the system prepares for, and record the brand profiles those channels belong to. You can also register accounts your brand owns or has licensed, with an authorisation note and supporting evidence, so the system knows which sources are genuinely yours. * **Transformative review consent** - An explicit opt-in before Media Radar will work with third-party media beyond research, with the responsibilities spelled out. Anything produced this way stays review-only with attribution attached, and it cannot be scheduled or published automatically. * **Media Pack review** - A review queue where you judge the picks the system would have made for your top three categories, including the proposed treatments and drafted copy, and confirm whether it chose well or was right to skip rather than produce filler. Reviewing here creates nothing: no cards, charts, drafts, schedules or posts. Only a workspace owner or admin can record a verdict, and a skip cannot be approved until you have inspected the candidates the system actually considered. ## Meme Intelligence A dedicated dashboard for meme-token workspaces, covering what is moving the token and what to post next. Most of it is not built yet. * **Meme Intelligence** (beta / in progress) - Shows how many posts your workspace has generated and gives you quick access to your token's external pages such as market screeners and listing sites, prompting you to add any that are missing. The headline panels for top reply packs, best-performing meme format and narrative momentum are labelled coming soon on the page itself, so you can see what is available today and what is not. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://thecontentforge.gitbook.io/thecontentforge-docs/feature-glossary/glossary/trends-and-media.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.