Module 11: CITIZEN ACTION CARD, Read Your Own Feed
How to Run the 15-Minute Convergence Audit on Any Platform You Already Use
CITIZEN ACTION CARD — Module 11
Shadow Patriots Action Library · Project Milk Carton
Pairs with Module 11: “Two Rooms, No Door, One Voice” — Competing Narratives & The Mirror Image
Module 11 is the first information-architecture module in this series. The citizen-action move on Module 11 is structurally different from the moves on Modules 3 through 10. The earlier modules trained you to read documents — Form 990s, Schedules I, L, R, and the statutes that governed each. Module 11 trains you to read your own feed.
The audit takes fifteen minutes. The product is screenshot-grade evidence of a convergence event — a single eight-word frame, published by two accounts that publicly hate each other, against the same target, on the same day, with no contact between them. The citizen who runs this audit once never reads a feed the same way again.
What a Convergence Audit Looks Like on Paper
A convergence audit produces five public-record fields per matching post. You need them to make the case structurally — to yourself, to a reporter, to a regulator, or to anyone who challenges your read.
Two posts with the same exact phrase, posted within 24 hours, by accounts of opposite political coding, with no message thread or shared employee between them, is the structural signature of convergence. You are not proving collusion. You are documenting that the algorithm produced the same frame in two opposing rooms within a single news cycle.
The 15-Minute Convergence Audit
Step 1 — Pick a phrase (2 minutes)
From the last week of your own feed, pick a single phrase, three to eight words long, that you remember encountering more than twice. The phrase should be specific enough that random people would not write it the same way by accident — a particular noun, a particular adjective, a particular construction that does not appear in everyday speech. Common phrases (“billionaire dark money,” “weapons of mass destruction,” “the deep state”) will not work for this audit because they are part of ordinary discourse. You want the unusual phrasing — the one that struck you as oddly specific the first time you saw it, and oddly familiar by the third time.
The candidate phrase is the seed of your audit.
Step 2 — Search the phrase in quotation marks (2 minutes)
On X, Bluesky, Threads, Truth Social, or whichever platform the phrase first appeared on, paste the phrase in quotation marks into the search bar. Quotation marks tell the search engine to return only posts containing the exact string, in the exact word order. Without quotation marks, the search returns paraphrases and related content; with quotation marks, you get only the precise frame.
Use the platform’s advanced search to set a 48-hour window centered on the date you remember first seeing it. Hit search.
Step 3 — Sort by date, find the earliest (3 minutes)
Sort the results from oldest to newest. Note the timestamp on the earliest matching post — date, hour, minute. Click through to the account that posted it.
On the account’s profile page, scan the bio (one or two sentences), the pinned post (top of the timeline), and the last 20 replies. Make a behavioral judgment about the account’s political coding. You are looking for unbroken pattern — does this account consistently defend figures associated with the political left, or with the political right, or neither? Three signals that agree across the bio, the pinned post, and recent replies is enough. The judgment does not have to be perfect; it has to be honest.
Write down on a notepad or in a phone note: - Earliest post timestamp - Account handle - Political coding (left / right / neither) - Follower count
Step 4 — Walk forward through the search results (5 minutes)
Look at the next post in the search results, and the next, and the next. Note the same five fields for each: timestamp, handle, political coding, follower count. Continue walking forward in chronological order. You are watching the cascade arrive in real time, on your own screen.
By the time you reach the fifth or sixth post, you will start to see the structure: a dense cluster of in-cluster amplification right after the first post (left-coded accounts re-posting a left-coded original, or vice versa), and then — at some hour mark — the cross-spectrum jump. A post in the opposing political cluster, using the same phrase verbatim. That is the moment convergence happened.
Step 5 — Document the convergence (3 minutes)
The defining marker of convergence — as opposed to ordinary in-cluster amplification — is the moment the phrase that started in one political cluster appears, in identical form, inside the opposing cluster within 24 hours.
Take screenshots of: - The earliest matching post (the first publication of the frame) - The first cross-spectrum post (the cluster jump)
Make sure both screenshots show the timestamp and the handle clearly. If the platform’s interface hides the timestamp on a screenshot, hover or tap to reveal the full date and time before capturing.
You now have a two-screenshot record of a real convergence event, observed on your own feed, on a target you chose, using nothing but a public platform’s own search tool. The product is a map. The map is not an accusation. The map is an observation that two political clusters that publicly hate each other produced the same framing on the same target on the same day, without coordination.
Score the Convergence
One point each — if the answer is yes for the phrase you searched.
The phrase appeared in two political clusters within 24 hours. (The cross-spectrum jump occurred quickly. Slower jumps are less suggestive of convergence; faster jumps are more.)
The two earliest publishers, one in each cluster, have a combined reach above one million followers. (Reach amplifies the structural impact. Smaller-reach jumps may be ordinary mimicry; large-reach jumps are the algorithm operating at scale.)
The two earliest publishers in the cross-spectrum jump have no public history of communication. (Mute history, replies that are hostile, or no replies at all between the two accounts is the structural signature of independent action.)
At least three additional accounts in each political cluster picked up the phrase within 48 hours. (The cascade structure — not a one-off — is what distinguishes convergence from coincidence.)
The phrase makes a specific factual or characterological claim about a named target. (Frames that name a target carry the structural significance the doctrine cares about. Generic political phrases that name no target are ordinary discourse, not convergence.)
Score bands
0–1 — coincidence or in-cluster amplification. Not a convergence event.
2–3 — possible convergence. Worth a closer look. Capture screenshots; check follower counts; note the timing.
4–5 — high-confidence convergence event. The phrase has crossed the spectrum, on a named target, with reach, without public communication between the cross-spectrum publishers, in a documented cascade. Save the screenshots. This is the structural signature.
The score is not a verdict on any specific operator. The score is a description of the event. Convergence events happen continuously, on platforms designed in part to produce them. The audit teaches you to see them. Seeing them is the literacy.
Two More Audits the Same Tools Unlock
The convergence audit teaches you to read your own feed for one specific structural pattern — a phrase moving across opposing political clusters within twenty-four hours. The same five-minute search-bar discipline opens two additional audits that are worth knowing about. Each takes about ten more minutes; each produces its own screenshot-grade evidence; each strengthens the citizen’s capacity to step outside the architecture and observe it from the outside.
The Designation Audit
Pick a person you have a strong opinion about — public figure, journalist, activist, candidate, anyone whose name comes pre-loaded with a verdict in your brain. Pick the single phrase your brain reaches for when you describe them: “extremist,” “grifter,” “terrorist sympathizer,” “fascist,” “communist,” “domestic terrorist,” “propagandist,” “useful idiot,” “election denier,” “denier” of any specific thing.
Now ask: who originally produced that designation? What is the methodology behind it? Can I read the criteria?
Open a search engine. Find the earliest authoritative use of the designation as applied to that specific person. Track it back to the producing organization — a nonprofit, a think tank, a journalism outlet, a watchlist, a platform trust-and-safety policy, a database. Find the organization’s published methodology for how the designation is assigned. Read the criteria. Audit the criteria against the specific person.
The minute you cannot find the methodology — or you can find it but cannot verify that the criteria were applied consistently to other people who fit the criteria but did not receive the designation — you are using a designation system whose internal mechanics are unauditable. You are outsourcing political judgment to an organization you have not met, using methodology you cannot read. That is structurally fine when you have decided to do it knowingly. It is structurally a problem when you have done it without realizing.
The designation audit’s product is not a verdict on the person. The designation audit’s product is a citizen who knows when they are using a designation versus making a judgment.
The Replacement Audit
Pick the same person from the designation audit. Now build your own designation, from public records, in fifteen minutes.
Open ProPublica Nonprofit Explorer for any (c)(3) or (c)(4) they run, sit on the board of, or are paid by. Open the FEC for any campaign contributions they have made or received. Open PACER for any federal court history. Open state corporate registries for any LLC or corporate filings they appear in. Open their LinkedIn for their declared affiliations. Open their own published writing for what they have publicly stated about themselves.
Build a one-page profile from those sources. Write down what you found. Note what is and is not in the public record. Note what is asserted versus what is documented.
Compare your one-page profile to the designation you started with. Where do they agree? Where do they diverge? What did the designation include that the public record does not support? What did the designation omit that the public record makes visible?
You have just performed the analytical work the designation system performs for the country every day. You have just earned the right to disagree with the designation, or to agree with it, on your own evidence. You did not need anyone else’s verdict. The records are public. You just had to read them.
The replacement audit’s product is the citizen who, the next time someone hands them a verdict, asks: what does the public record actually show? That citizen carries a tool the entire designation architecture was not built to defeat.
Quick Reference — Platform Tools
The platform’s own search is the primary tool. Archive.today is the secondary tool. When you find the cross-spectrum jump, snapshot both posts at archive.today before posting screenshots anywhere. Posts can be deleted; archive.today snapshots persist.
A Reminder About What This Card Is NOT
This card is not a guide to identifying coordinated inauthentic behavior. Convergence is not coordination. The whole point of the audit is to teach you to see the effect the regulatory framework was designed to detect, produced without the coordination the framework requires as evidence. If you complete the audit and find a convergence event, you have not proven anyone broke the law. You have observed that the algorithm produced an effect the law cannot reach.
This card is also not a guide to outing, doxing, or pile-on coordination against the accounts you find. The audit is a literacy exercise. The accounts you identify are not your targets. They are the operators of the architecture you are mapping. Naming them publicly, sharing their handles widely, or organizing reply campaigns against them would make you a participant in the same information operation you are learning to step outside of. The Shadow Patriots do not expose. We map. The map is for you.
The audit’s product is the citizen who runs it — the citizen who, the next time a phrase moves through their feed, will instinctively search the phrase, sort by date, and check the political coding of the earliest publishers. The audit becomes a reflex. The reflex becomes a literacy. The literacy becomes the only thing the algorithm cannot route around.
Every citizen who runs the audit once steps outside the operation. Every citizen who steps outside the operation makes the architecture one step more legible.
Shadow Patriots Action Library · Module 11 · Project Milk Carton · 501(c)(3) · EIN 33-1323547