Verified Poster Pipeline Explorer
Run the Verified Poster Pipeline Explorer MicroSim Fullscreen
About This MicroSim
A fact-verified poster is a static poster whose every numeric claim has been traced to a cited source and a quoted passage before any pixel is rendered. The generator skill's verified-infographic guide organizes the work as eight phases plus a ninth overlay step. This explorer draws the route as a chain that reads left to right and wraps like lines of text:
- Claim plan (
01-claim-plan.yaml) - Source discovery (source records go into the verification report)
- Verification (
02-verification-report.md) - User checkpoint (approves the report; writes nothing new)
- Layout spec (
03-layout-spec.yaml) - Prompt assembly (
04-image-prompt.md) - Render (
poster.png) - Render audit (
sources.md) - Overlay (
data.jsonandmain.html)
Two steps are colored. Step 4, the User checkpoint (orange), is the mandatory human checkpoint: nothing is laid out or rendered until the requester approves the claim set. Step 7, Render (purple), is the only step that calls the image model. A dashed red arrow from step 8 back to step 7 is the retry loop: when the audit finds drift, the poster is rendered again, up to three attempts before escalating to the user. A green bracket marks steps 1 to 4 as reusable on their own, because a verified claim set can feed an interactive chart or table with no poster at all.
Click a step to read a two-sentence description, the file it writes, and the question "What goes wrong if this step is skipped?" Hover an arrow to see what passes along it. The Skip a step menu greys out one step and shows the failure the guide anticipates, such as unverified numbers reaching the prompt. The Sequence quiz shuffles the unnumbered steps: click them in order, then pick the checkpoint and the image-model step.
Learning objective: The learner will sequence the nine steps of the verified poster route, and will explain which step is the mandatory human checkpoint and which step alone calls the image model.
Bloom's taxonomy level: Understand (verb: sequence)
How to Use
- Read the chain from step 1 to step 9. Hover each arrow to see what the next step receives.
- Click a step. Before pressing Show the answer, predict what goes wrong if it is skipped.
- Choose a step in Skip a step to grey it out and read the failure it would cause. Choose (none) to restore the chain.
- Press Sequence quiz. Click the nine shuffled steps in order; out-of-order clicks are counted. Then click the mandatory checkpoint and the image-model step. Press Explore to return.
When the page is narrower than 600 pixels, the information panel moves below the chain.
Iframe Embed Code
You can add this MicroSim to any web page by adding this to your HTML:
1 2 3 4 | |
Lesson Plan
Audience
Teachers, instructional designers, learning-technology developers and learning-analytics practitioners (college undergraduate and professional development).
Duration
15 minutes
Prerequisites
- The Fact-Verified Poster section of Chapter 11
- Grid overlays and comparison posters from Chapter 10
Activities
- Trace (4 min): Hover every forward arrow and write one line naming what each step hands to the next.
- Predict and skip (6 min): For steps 3, 4, 6 and 8, predict the failure before choosing the step in Skip a step. Compare your prediction with the failure shown.
- Sequence (5 min): Complete the Sequence quiz. Record out-of-order clicks and whether you found the checkpoint and the image-model step on the first try.
Assessment
- Out-of-order clicks in the Sequence quiz (zero is the goal).
- In one or two sentences, the learner explains why step 4 must come before step 5 and why no step other than step 7 calls the image model.
- Given a poster with a misprinted number, the learner names the step that should have caught it (step 8) and the step that the retry loop returns to (step 7).
References
- Fact-checking - Wikipedia - Background on verifying factual claims against sources before publication.
- Infographic - Wikipedia - Posters that present information and data visually.
- Text-to-image model - Wikipedia - The kind of model called once, in step 7.
- vis-network documentation - The network library used to draw the pipeline.