Executive Overview
Rhetoric Audit is a forensic media analysis platform delivered as a web application and Chrome extension. It enables analysts, journalists, policy researchers, and critical readers to evaluate the rhetorical integrity of media — from a single article to multi-source intelligence briefs.
Purpose
This document defines the primary product use cases based on the current implementation and supporting documentation.
- Product handoff
- Stakeholder communication
- Development planning
- QA and UAT traceability
- Proposal and client documentation
Product Scope
Rhetoric Audit is a web platform plus Chrome extension for forensic media analysis, supporting two major analytical workflows.
Single-article rhetorical analysis
Analyze a specific article and receive structured metrics: ideological framing, manipulation risk, fallacy density, evidence validity, transparency, emotional resonance, and narrative archetype.
Multi-source Intelligence Brief
Submit search terms and selected sources (X, Reddit, News, YouTube, Web, Financial) and receive a synthesized brief covering narratives, authenticity, emotion, risk, and cross-platform signal patterns.
3. In Scope
- Marketing site and onboarding journey
- Account creation and login
- Chrome extension assisted article analysis
- Article analysis result review
- Dashboard and analytics consumption
- Scan history search and filtering
- Intelligence Brief generation and retrieval
- Credit purchase and balance management
- Notification preference management
- Account deletion flow
4. Out of Scope
- Dedicated internal admin panel
- Enterprise team management
- Subscription billing
- Role-based multi-user workspaces
- Moderation and compliance backoffice workflows
Primary Actors
| Actor | Description |
|---|---|
| Visitor | A public user browsing the landing page and evaluating the product. |
| Registered User | A user with an account who can log in and access protected features. |
| Analyst User | The authenticated user performing scans, reviewing results, and generating briefs. |
| Billing User | The authenticated user purchasing additional credits. |
| Support / Admin System | Operational owner responsible for retained data, billing events, and service management. |
Supporting External Systems
Key Business Rules
- A user must be authenticated to use protected analysis APIs.
- Single article analysis consumes 1 credit.
- Intelligence Brief generation consumes 2 credits.
- Low-credit alerts trigger at 3 credits or below when the user has opted in.
- Intelligence Brief generation is rate-limited per user.
- Shared analysis pages expose a limited public summary, not the full private dashboard record.
- Credit purchases are one-time transactions, not subscriptions.
- Account deletion removes the user account; historical analysis data is retained in the backend for operational reasons under the current implementation.
Use Case Map
- Browse product site
- Create account / Log in
- Install Chrome extension
- Run article analysis
- View dashboard and analytics
- Search scan history
- Share analysis summary
- Generate Intelligence Brief
- Review saved Intelligence Brief
- Manage profile and notifications
- Delete account
- Receive operational emails
- Purchase credits
Use Case Catalog
| ID | Use Case | Primary Actor | Outcome |
|---|---|---|---|
| UC-01 | Browse product and understand value proposition | Visitor | User understands product capabilities and target use cases. |
| UC-02 | Register or log in | Visitor / Registered User | User gains authenticated access to the platform. |
| UC-03 | Install Chrome extension | Visitor / Registered User | User can launch article analysis from Chrome. |
| UC-04 | Run single-article rhetorical analysis | Analyst User | Article is analyzed and stored as a scan result. |
| UC-05 | View dashboard and analytics | Analyst User | User monitors activity, performance, and scan trends. |
| UC-06 | Search and review scan history | Analyst User | User finds and revisits previous analyses. |
| UC-07 | Share an analysis summary | Analyst User | User shares a compact result view externally. |
| UC-08 | Generate an Intelligence Brief | Analyst User | Multi-source narrative brief is produced and stored. |
| UC-09 | Review Intelligence Brief history | Analyst User | User retrieves previously generated briefs. |
| UC-10 | Purchase credits | Billing User | User receives additional credits for continued usage. |
| UC-11 | Manage profile and notification settings | Registered User | User updates preferences and profile information. |
| UC-12 | Delete account | Registered User | User account is removed from active service access. |
| UC-13 | Receive transactional notifications | Registered User | User is informed about low credits or scan completion. |
Detailed Use Cases
Browse Product and Understand Value Proposition
Preconditions
- Public website is available.
Main Success Flow
- 1Visitor lands on the homepage.
- 2Visitor reviews core capabilities such as bias spectrum mapping, propaganda index, and strategic silence.
- 3Visitor reviews intended audiences such as researchers, journalists, policy analysts, and critical readers.
- 4Visitor reviews the How It Works flow.
- 5Visitor chooses to install the extension or create an account.
Alternate Flows
- Visitor leaves without creating an account.
- Visitor proceeds directly to Chrome extension installation.
Postconditions
- Visitor has enough product understanding to continue to onboarding or evaluation.
Register or Log In
Preconditions
- User has internet access.
- Authentication service is available.
Main Success Flow
- 1User opens the sign-up or login form.
- 2User authenticates using email/password or Google OAuth.
- 3System creates or validates the session.
- 4System redirects the user to the dashboard or sign-up success flow.
Alternate Flows
- Password mismatch prevents sign-up.
- Invalid credentials block login.
- OAuth callback fails and the user is returned to an error state.
Postconditions
- User has an authenticated session and can access protected routes.
Install Chrome Extension
Preconditions
- User is using Chrome or a compatible browser.
Main Success Flow
- 1User opens the Chrome Web Store listing.
- 2User installs the extension.
- 3User browses to an article.
- 4User launches the extension from the browser toolbar.
Alternate Flows
- User does not complete installation.
- Browser does not support the extension.
Postconditions
- Extension is available for article analysis.
Run Single-Article Rhetorical Analysis
Preconditions
- User is authenticated.
- User has at least 1 credit.
- LLM analysis service is available.
Main Success Flow
- 1User provides an article URL or raw article text.
- 2System validates the request.
- 3System checks the user's available credits.
- 4System performs FME analysis.
- 5System generates the structured result set.
- 6System stores the analysis.
- 7System deducts 1 credit.
- 8System displays the result to the user.
- 9System may email a scan-complete summary if that preference is enabled.
Primary Outputs
Alternate Flows
- User is unauthenticated and is redirected to login.
- User has insufficient credits and is prompted to top up.
- Analysis fails due to upstream model or service error.
- Request is malformed and validation fails.
Postconditions
- Scan result exists in the user's history if processing succeeds.
View Dashboard and Analytics
Preconditions
- User is authenticated.
- At least one scan may exist for meaningful analytics.
Main Success Flow
- 1User opens the dashboard.
- 2System displays summary KPIs such as total scans, average evidence score, average source credibility, credits remaining, manipulation risk, transparency, and fallacy rate.
- 3User reviews recent scan activity and risk charts.
- 4User navigates to analytics.
- 5System displays scan activity, ideology distribution, manipulation risk distribution, and time-based trends.
Alternate Flows
- User has no scans and sees empty-state messaging.
Postconditions
- User understands personal analysis patterns and account usage status.
Search and Review Scan History
Preconditions
- User is authenticated.
- Scan history data exists or the system can return an empty list.
Main Success Flow
- 1User opens the Scan History view.
- 2System lists prior scans with article title, ideology, manipulation risk, fallacy, evidence, depth, and date.
- 3User searches by text or filters by ideology and date range.
- 4System refreshes the result set.
- 5User reviews a previous record and may open the original article URL.
Alternate Flows
- No scans match the filters.
- Pagination is required because the result set is large.
Postconditions
- User can locate and review historic analyses efficiently.
Share an Analysis Summary
Preconditions
- An analysis result exists.
Main Success Flow
- 1User initiates sharing from an existing analysis.
- 2System constructs a shareable summary view using selected result fields.
- 3Recipient opens the shared analysis page.
- 4Recipient sees the title and summary metrics such as manipulation risk, fallacy rate, evidence, and narrative archetype.
Alternate Flows
- Shared link is incomplete and renders a partial view.
Postconditions
- A limited analysis summary can be viewed outside the private dashboard.
Generate an Intelligence Brief
Preconditions
- User is authenticated.
- User has at least 2 credits.
- Edge function and LLM services are available.
Main Success Flow
- 1User opens the Intelligence Brief module.
- 2User enters one or more search terms.
- 3User selects sources such as X, Reddit, News, YouTube, Web, and Financial.
- 4User submits the request.
- 5System validates the payload and enforces rate limits.
- 6System checks available credits.
- 7System calls the edge function to gather normalized signals from selected sources.
- 8System sends the signal set to the LLM analysis pipeline.
- 9System validates and stores the resulting intelligence brief.
- 10System deducts 2 credits and logs the transaction.
- 11System presents the results dashboard and history access.
Primary Outputs
Alternate Flows
- User lacks credits and receives an insufficient-credit response.
- Per-user rate limit is exceeded.
- Some sources fail, but partial results are still accepted when allowed.
- No usable signal set is returned and the brief cannot be generated.
Postconditions
- Intelligence Brief is stored and retrievable from history if the run succeeds.
Review Intelligence Brief History
Preconditions
- User is authenticated.
Main Success Flow
- 1User opens the Intelligence Brief history tab.
- 2System lists previous brief records.
- 3User selects a brief.
- 4System loads the saved brief results into the results view.
Alternate Flows
- No previous briefs exist.
Postconditions
- User can continue analysis or presentation work using saved brief outputs.
Purchase Credits
Preconditions
- User is authenticated.
- Stripe checkout service is available.
Main Success Flow
- 1User opens the Billing page.
- 2System displays available credit packs.
- 3User selects a pack.
- 4System creates a Stripe Checkout session.
- 5User completes payment in Stripe.
- 6System records the billing event and applies credits to the account.
- 7User returns to the application with an updated balance.
Alternate Flows
- User cancels checkout.
- Stripe session creation fails.
- Payment succeeds but post-payment credit application is delayed and requires reconciliation.
Postconditions
- User has additional credits if payment succeeds.
Manage Profile and Notification Settings
Preconditions
- User is authenticated.
Main Success Flow
- 1User opens the Settings page.
- 2User updates display name.
- 3User enables or disables low-credit notifications.
- 4User enables or disables scan-complete notifications.
- 5User saves changes.
- 6System updates the profile record.
Alternate Flows
- Save operation fails and the user is shown an error.
Postconditions
- Profile and notification preferences are updated.
Delete Account
Preconditions
- User is authenticated.
- User confirms intent by entering the required confirmation text.
Main Success Flow
- 1User opens the delete account confirmation modal.
- 2System explains the consequences.
- 3User types DELETE to confirm.
- 4System calls the delete-account endpoint.
- 5Backend reassigns retained analysis and invoice records according to current operational implementation.
- 6Backend removes the user profile.
- 7Backend deletes the authentication record.
- 8User is signed out and redirected away from the protected area.
Alternate Flows
- User closes the modal without confirming.
- Confirmation text is incorrect.
- Backend deletion partially succeeds and requires support follow-up.
Postconditions
- User account no longer exists for login purposes.
Receive Transactional Notifications
Preconditions
- User has opted in to the relevant notification type.
- Email service is configured and available.
Main Success Flow
- 1User performs activity that changes account or scan status.
- 2System evaluates notification rules.
- 3If the low-credit threshold or scan-complete condition is met, the system generates an email.
- 4Email is sent to the user's registered address.
Alternate Flows
- Email provider is unavailable.
- Notification preference is disabled.
- Email configuration is missing.
Postconditions
- User receives informational email when conditions are met and delivery succeeds.
Preconditions Summary
| Capability | Auth Required | Credit Required | External Dependency |
|---|---|---|---|
| Browse landing page | No | — | Website only |
| Install extension | No | — | Chrome Web Store |
| Single article analysis | Yes | 1 | LLM service |
| Dashboard and analytics | Yes | — | Database |
| Scan history | Yes | — | Database |
| Intelligence Brief | Yes | 2 | Edge function + LLM |
| Buy credits | Yes | — | Stripe |
| Settings update | Yes | — | Database |
| Account deletion | Yes | — | Auth + database |
| Email notifications | Yes | — | Resend |
Success Criteria
- Visitors can understand the offer and move into onboarding.
- Users can authenticate successfully with email/password or Google.
- Article scans return structured forensic results.
- Scan history and analytics reflect stored usage.
- Intelligence Briefs can be generated, stored, and reopened.
- The credit model enforces access correctly.
- Billing can add credits through Stripe.
- Settings changes persist correctly.
- Account deletion removes login access.
- Transactional emails send when enabled and conditions are met.
Risks and Notes
- The repository shows a transition from extension-first delivery to a broader SaaS platform; older documentation may describe only part of the current scope.
- Account deletion UX describes anonymized retention while backend logic currently reassigns analysis records to an admin-owned target. This should be reconciled in formal policy documentation.
- Intelligence Brief functionality is mature in technical documentation but still labeled BETA in the dashboard navigation; stakeholder-facing messaging should clarify release status.
Looking for current documentation?
This is the v1.0 archive. See v2.0 for updated FME V19.1 capabilities, V3 ambient extension, span-level analysis, and factual grounding.
View v2.0 Docs →