Daily Brief — 2026-06-03

> Generated: 2026-06-04 07:01 > Model: gemini-2.5-flash > Tokens: 57,879 in / 6,727 out > Cost: $0.0127 > Preset: small_team

---

CRITICAL SYSTEM ALERT: AIOS DATA COLLECTION FAILURES

Sondra, the AIOS system health is currently RED. Three critical scheduled tasks failed yesterday: AIOS Daily Brief, AIOS Dashboard Push, and AIOS IntelOS Daily Collection. This means the data presented in this brief may be stale or incomplete, and your dashboard may not have been updated. This requires immediate investigation to ensure the integrity of your intelligence system.

The Day in Brief

Your AIOS system is currently experiencing critical failures, with daily data collection and dashboard pushes failing yesterday. This means the intelligence you rely on may be stale, and resolving this should be your top priority to restore full operational visibility.

Despite the system issues, significant progress was made on the doTERRA Monthly Drop Club, with all 175 scripts being regenerated to a new gold standard, ensuring fresh, non-repetitive content. This is a direct outcome of the Agent Opus Challenge, which continues to be a high-priority focus, now extended by one week due to credit-migration bugs.

In your weekly Stars training, you demonstrated Claude's powerful skill-building capabilities, directly addressing Tina Alexander's desire to create websites for clients. This session highlighted the value of custom skills for automating content creation and design, a core offering for your community. You also engaged with a doTERRA-focused webinar from Mandy Sommers, which presented a structured reactivation program, a relevant offering given your doTERRA income stream.

On a personal strategic front, your Ikigai reflection provided profound clarity on your core identity as a "create and teach" entrepreneur, with AI as the current material. This introspection, however, also surfaced the "sustain" pillar as your lowest score, underscoring the ongoing challenge of Activator drift and the need to finish existing projects before starting new ones. The upcoming July 1 Stars Premium price increase presents a clear opportunity to drive annual sign-ups at the current rate.

Key Signals

⚠️ System Health RED: All three critical AIOS scheduled tasks (Daily Brief, Dashboard Push, IntelOS Daily Collection) failed yesterday, indicating stale or missing data. 🔥 Monthly Drop Club Scripts Rerun: All 175 doTERRA scripts were regenerated to a new "gold standard," ensuring fresh, non-repetitive content for Agent Opus. 📊 Stars MRR Growth: Stripe MRR increased by $6.00 to $95.75/mo, and active subscriptions grew by 1 to 19 members this week. 💡 Claude Skills Demo Impact: Your Stars training on creating and using Claude skills directly addressed member needs (Tina Alexander's website goal), showcasing a powerful value proposition. 📌 Ikigai Clarity: Deep personal reflection confirmed "create and teach" as your core identity, with "sustainability" (score 69) identified as the key area needing focus. 🔥 Romania Prospect Progress: The live Romania Business in a Box prospect, Azaleea Zaman, is advancing, with a cost breakdown pending. ⚠️ Activator Drift Warning: Your engagement with Mandy Sommers' doTERRA program, while relevant to your runway, highlights the ongoing risk of new projects competing with core AI business priorities. 💡 Stars Price Increase Opportunity: The July 1 price increase for Stars Premium (to $47/mo and $297/yr) creates a strong incentive to push annual sign-ups at the current rate for the next four weeks. 📌 Storefront Bottleneck: The "Everything Sondra Sells" storefront remains "NOT BUILT," leaving 50+ apps and GPTs unmonetized despite being ready assets. 📊 YouTube Subscribers Up: Your YouTube channel gained 60 subscribers, reaching 1,060, indicating continued audience growth.

Metrics Analysis

Overall System Health: The most critical signal is the RED system health. The failure of the AIOS Daily Brief, AIOS Dashboard Push, and AIOS IntelOS Daily Collection tasks means that the data presented here may be stale, and the dashboard you rely on for real-time insights is not updating. This directly impacts your ability to make informed, data-driven decisions and should be addressed immediately.

Stripe Revenue: Your Stripe MRR shows positive movement, increasing by $6.00 to $95.75/mo this week, above last week's $89.75. Active subscriptions also grew by 1, reaching 19. Revenue MTD stands at $799.0, with 4 new subscriptions this month and 1 cancellation, resulting in a 4.5% churn rate. This indicates slow but steady growth in your recurring revenue streams, primarily from AmplifAI Video Studio and potentially some Stars Premium members billed via Stripe. The upcoming Stars price increase could significantly boost this if leveraged effectively.

Skool Stars Premium: While the current-data.md notes 19 paying members (16 premium + 3 VIP) as of June 1, the income-snapshot.md indicates a total Stripe MRR of $95.75/mo across 19 active subscriptions. This suggests that the Skool Stars MRR of ~$207/mo mentioned in the wiki is likely outdated and needs recomputation based on current Stripe data. The actual recurring revenue from your AI education brand is currently lower than the stated target of $1,000+/mo, reinforcing the "gap" problem you are solving.

AmplifAI Video Studio: This product contributes to your overall Stripe MRR. The income-snapshot.md shows 18 paying subscribers across various tiers ($3, $5, $6/mo, $24, $57/yr), contributing to the $95.75 total MRR. The progress on regenerating 175 doTERRA scripts for the Monthly Drop Club is a significant operational win that will enhance the value and content pipeline for these subscribers.

YouTube Channel: Your YouTube channel continues to grow, reaching 1,060 subscribers, up from 1,000 previously. This is a positive sign for awareness, your first funnel stage. However, the "Videos Published (30d)" is only 1, and "Views (30d)" is 18, which is below average for a channel of this size. This suggests that while your audience is expanding, your recent content output on YouTube is low, aligning with the decision to revamp the channel via Sandra Lee's program. The top videos ("AI Just Copied This Website Perfectly," "Claude vs ChatGPT") indicate strong interest in practical AI application.

Skool Affiliate Earnings: Lifetime affiliate earnings increased by $53.20 this week, reaching $2,260.39. This consistent, albeit smaller, revenue stream is a positive indicator of your influence within the broader AI and entrepreneurial communities.

Meeting Highlights

Call 1: Impromptu Zoom Meeting with Molly Mahoney & Lindsay York (2026-06-03)

• Notable Signals: This was a general webinar focused on tracking information and a "secret workbook" from Lindsay York. Sondra was an attendee, not a presenter. No direct action items or decisions for Amplif-AI.

Call 2: STARS - Weekly Training with Tina Alexander & Carolyn Hyatt (2026-06-03)

• Key Decisions Made:
• Sondra committed to fully launching Claude training next week for Stars members.
• Sondra will provide some pre-built Claude skills to paid members and teach them how to create their own.
• Action Items:
• [Sondra] Start full Claude training for Stars members next week.
• [Sondra] Prepare to provide pre-built Claude skills to Stars members.
• [Sondra] Teach Stars members how to create their own Claude skills.
• [Sondra] Upload the replay of this training to the community area.
• [Sondra] Connect the dots for challenges in the community, specifically linking the "paying to profit" challenge to relevant pages.
• Notable Signals:
• Tina Alexander expressed a clear desire to build "beautiful websites that people would pay me for," seeking vibe coding and Claude integration. This directly aligns with your expertise and the Claude Lab training priority.
• Carolyn Hyatt attempted to migrate ChatGPT memory to Claude but failed, indicating a common user challenge that your training can address.
• Sondra successfully demonstrated Claude's skill creation and usage, generating 10 scroll-stopping Facebook angles and showing Gamma integration for presentations.
• Sondra highlighted the "Skill Security Scan" as a crucial skill to provide to members for vetting third-party skills.
• Sara Dependencies: Sara is explicitly on hold with a full-time job, and all previous collaboration is paused. Therefore, there are no open Sara commitments from this meeting.

Call 3: Impromptu Zoom Meeting with Mandy Sommers- Sommers School (2026-06-03)

• Key Decisions Made: This was a doTERRA-focused webinar on reactivating customers and builders. Sondra was an attendee, not a presenter. No direct decisions for Amplif-AI.
• Notable Signals:
• Mandy Sommers launched a "Reactivated doTERRA Summer School," a 12-week program priced at $197, starting June 15, focused on reactivating cold lists and building systems. This is a direct competitor or complementary program within the doTERRA space, relevant to Sondra's financial runway.
• The presentation emphasized mindset, overcoming guilt, and consistent action for doTERRA builders, echoing some of Sondra's own Ikigai reflections on "sustainability."
• Sara Dependencies: None.

Strategic Recommendations

1. Concerns:

• System Health Critical: The RED status of your AIOS scheduled tasks is a severe concern. Without reliable data collection and dashboard updates, your ability to monitor business performance, identify trends, and make informed decisions is compromised. This must be the absolute top priority for resolution.
• Activator Drift Persists: While the Agent Opus Challenge is active and progressing, the core AI business launches (AI Asset Weekend Builder, AI Marketing Team) remain "pending" or "in design." Your engagement with external doTERRA programs (Mandy Sommers) and new tool explorations (Gamma, Claude skills) risks further diluting focus from these critical, revenue-generating priorities, aligning with your identified "Activator drift" risk pattern.
• Storefront Monetization Gap: The "Everything Sondra Sells" storefront is still "NOT BUILT," leaving a significant inventory of 50+ apps and GPTs unmonetized. This represents a substantial missed revenue opportunity and a bottleneck in your product ecosystem.

1. Opportunities:

• Stars Price Increase Leverage: The confirmed July 1 price increase for Stars Premium (from $33/mo to $47/mo and $197/yr to $297/yr) provides a powerful, time-sensitive incentive. Aggressively promote annual sign-ups at the current lower rate over the next four weeks to maximize recurring revenue before the jump.
• Claude Skill Training as a Stars Pillar: Your successful demo of Claude skill creation and application in the Stars training directly addresses member needs and aligns with the "Claude Lab Training Inside Stars" priority. This is a high-value offering that can drive Stars membership and engagement, especially for members like Tina seeking to build websites.
• Romania Business in a Box: The active prospect with Azaleea Zaman for the doTERRA international partner program is a concrete step towards diversifying your doTERRA income stream and leveraging your Agent Opus capabilities for a franchise-like model. This is a high-potential opportunity for immediate follow-through.

1. Quick SWOT:

• Strength: Sondra's exceptional ability to simplify complex AI tools and teach practical application, as demonstrated in the Stars training, is a core differentiator.
• Weakness: The persistent Activator drift and the unbuilt "Everything Sondra Sells" storefront are delaying monetization of built assets and fragmenting focus.
• Opportunity: Capitalize on the upcoming Stars price increase and the proven demand for Claude skill training to accelerate recurring revenue growth.
• Threat: Critical AIOS system failures pose an immediate threat to data integrity and operational visibility, potentially undermining strategic decision-making.

Action Items

1. [Sondra] Resolve AIOS System Health Failures: Immediately investigate and resolve the RED status of scheduled_tasks (AIOS Daily Brief, AIOS Dashboard Push, AIOS IntelOS Daily Collection) to restore data integrity.
2. [Sondra] Drive Stars Annual Sign-ups: Actively promote the Stars Premium annual membership at the current $197/yr rate before the July 1 price increase to maximize conversions.
3. [Sondra] Prioritize Storefront Decision & Build: Make an immediate decision on the platform (Studio Market app vs. GHL page) and select the first 10 products to list for the "Everything Sondra Sells" storefront.
4. [Sondra] Advance AI Asset Weekend Builder: Focus on building the 3 flagship demo assets, with the Testimonial Vault as the primary example, to prepare for the next product launch.
5. [Sondra] Conduct AI Marketing Team GPT Security Testing: Initiate prompt injection tests with Claude for the AI Marketing Team GPT suite before public launch.
6. [Sondra] Build DFY Video Service Landing Page: Create the landing page for the Done-For-You Video Service ($49/mo or $99/mo) as the next step in its launch.
7. [Sondra] Continue Claude Lab Training: Maintain the bi-weekly Claude-focused sessions inside Stars, emphasizing practical skill creation and application, such as website building, to meet member demand.
8. [Sondra] Follow Up on Romania Prospect: Send the cost breakdown via WhatsApp to Azaleea Zaman for the AmplifAI Video Studio "Biz in a Box" opportunity and schedule remaining presentations.
9. [Sondra] Upload Stars Training Replay: Post the replay of the June 3 Stars training to the community area and connect the "paying to profit" challenge dots for easier member navigation.
10. [Sondra] Revisit 2026 Subscription Audit: Schedule a dedicated session to complete the 2026 subscription audit to gain full clarity on monthly expenses and identify non-ROI subscriptions for potential cancellation.

Your Open Questions

• Monthly Drop Club Script Pipeline:
• Scripts need full rerun (URL only): This requires a Claude Code session to confirm which scripts you manually rewrote, then rerun the remaining 175 scripts using only the URL as source, adhering to the new gold standard.
• Excel file locked: You must close and reopen the ~$FINAL doTERRA_All_Products_2026_with_scripts.xlsx file to remove the lock and allow changes to save.
• Ikigai — 3 unanswered questions: These are deep personal reflections for you to sit with: "AI or just create-and-teach?", "YouTube gap: foundation or publishing fear?", and "Shared-cost business: real desire or sunk-cost rescue?". No immediate action is required from Claude.
• Financial / Billing - Anthropic API billing change: You need to check your Anthropic billing dashboard and account settings to verify if programmatic (API) usage now bills separately from your $200/mo Claude Max plan.
• Misconfigured remote in PROJECTS/Pushten website/amplifai-template-studio/: This requires a Claude Code session to check if the origin remote pointing to AmplifAI-Your-Business/test-sondra-blog-website.git is an intentional clone or a stale misconfiguration.
• Base44 Blog Template — Productization Open Items:
• Apps 2 and 3 not yet replicated: This requires a Claude Code session to apply usePageSeo hook, dashboard toggles, and meta description fixes.
• Prerender.io vs Vercel migration vs marketing reposition decision pending: This is a strategic decision for you regarding the delivery of full AI-search visibility.
• og:image default still Base44 logo: This requires a Claude Code session to locate and update the platform-wide default OG image setting.
• 0 Content Types and 0 Niches defined on master: This requires a Claude Code session to define these, which is necessary for the pSEO Generate batch system.
• Owner Playbook v2 acceptance: You need to visually review the v2 PDF and explicitly accept or reject it.
• Customer-facing Owner Playbook needs updating: This requires a Claude Code session to update the playbook to reflect the verified state and honest Base44 limitations.
• Publish UX wall: This is a decision for you: either document the non-obvious "STATUS dropdown → Published" or send a Base44 editor prompt to replace it with [Save Draft] [Publish] buttons.
• AI Write Post voice quality: This requires a Claude Code session for voice-training intervention or clear "starting draft only" positioning before selling.
• Blog Owner Playbook - Visual review of v2 PDF: You need to open the v2 PDF in Edge, click 3-5 random Lookup Table rows to confirm hyperlinks, and skim the Site Config crops to verify accuracy.
• SEO Blog Website Template — Productization:
• Modal readability fix verification: You need to visually verify the fix for the .admin-white-card className in AdminModal.jsx after the most recent deploy.
• pSEO + indexing comprehensive fix deployment status: You need to verify the end-to-end functionality of the multi-feature prompt sent to all 3 apps.
• Master Base44 app selection: You need to select which of the three Base44 apps will serve as the production master for cloning to new buyers.
• Pricing tier formal confirmation: You need to formally accept or adjust the proposed pricing tiers for the website service.
• First buyer / launch path: You need to pick the first client target for this service.
• Infrastructure - Wrangler CLI not authenticated locally: You need to run wrangler login to authenticate the Wrangler CLI for Cloudflare operations.
• Digital Home Backend — Gemini swap + UI gaps:
• Gemini output quality vs Claude: You need to read the first draft article in the dashboard ARTICLES tab ("How to Use AI to Write a Weekly Newsletter in 15 Minutes") and judge its quality against your brand voice.
• Where was the original unauthorized Claude swap Sondra remembered?: This requires a Claude Code session to check other projects (Pushten website, Amplif-AI Template Studio, Coach Blog Template, BraveBrand intelligence app).
• Backend dashboard UI gaps: This requires a dedicated Claude Code session for UI improvements (Add Topic/Run Trend Scan/Write Article buttons, sidebar placeholders, dark theme, sidebar scroll bug).
• No POST handler on /api/content-calendar: This requires a Claude Code session to add a POST handler to enable the future Add Topic button.
• package-lock.json has 122 unrelated deletions: This requires a Claude Code session to verify or discard these changes in the backend repo working tree.
• Template-Blog-website Migration:
• Guessed column types in 0001_initial_schema.sql: Monitor for errors on specific admin pages; if they misbehave, a Claude Code session will be needed for column-type adjustments.
• Phase 4 or 5 next?: This is a strategic decision for you regarding the next phase of migration (backend functions + Resend vs. AI content via Claude API).
• Phase 4 — Backend functions still needed: This requires a Claude Code session to build publishScheduledPosts, generateSitemap, and weeklyReport using Supabase Edge Functions + Resend.
• Phase 5 — AI content generation pending: This requires a Claude Code session to integrate the Anthropic Claude API into AdminGenerate.jsx and AdminNewPost.jsx.
• Phase 6 — Media uploads + Supabase Storage not yet configured: This requires a Claude Code session to configure media uploads and Supabase Storage.
• GitHub auto-deploy to Vercel not connected: You need to connect the Vercel GitHub app to the AmplifAI-Your-Business org for auto-deploys.
• Custom domain for template-blog-website.vercel.app: You need to name and configure a custom domain for this project.
• Resend account for Phase 4 emails: You need to decide whether to reuse Digital Home's Resend account or create a separate one.
• TweetGod Supabase restore timing: You need to decide when the TweetGod bot goes live and needs its Supabase restored.
• Amplif-AI Template Studio:
• Purchase links: what payment mechanism?: You need to decide on the payment mechanism (Stripe direct links, GHL checkout, or ThriveCart) before the Phase 2 build.
• Custom domain for template studio?: You need to decide if amplifai-template-studio.vercel.app gets templates.amplifaiyourbusiness.com or its own domain.
• GitHub connection to Vercel not linked: You need to connect the Vercel GitHub app to the AmplifAI-Your-Business org for auto-deploys.
• TweetGod Supabase paused: You need to decide when the TweetGod bot goes live and needs its Supabase restored.
• Digital Home Backend - Frontend pages not yet branded: This requires a Claude Code session to generate a redesign prompt using BraveBrand Prompt Builder and apply it to the Digital Home frontend project.
• Brand Identity - context/brand-detail.md not yet updated: This requires a Claude Code session to load brand-detail and apply the updates from the May 2026 ICA and Brand Identity documents.
• GEO + Reputation Agency - Service domain strategy: You need to decide whether the GEO/reputation service will live on sondraverva.com or a separate brand domain.
• GEO + Reputation Agency - Self-serve app brand name: You need to decide on a brand name ("GEO Score" / "AI Visibility Score" / Sondra-branded) before the Phase 4 build.
• GEO + Reputation Agency - Jonathan Mast's GEO app: You need to research what this app specifically does and if it's competitive or complementary.
• Digital Home (SondraVerva.com):
• brand_context Supabase rows not yet added: This requires a Claude Code session to add the three necessary rows to the brand_context table.
• Pages not yet branded: This requires a Claude Code session to generate a redesign prompt using BraveBrand Prompt Builder and apply it to the Digital Home project.
• GitHub Actions need SITE_URL variable: You need to deploy the backend and set the SITE_URL repo variable in GitHub Settings → Secrets → Variables tab.
• Backend setup pending: This requires a Claude Code session to clone lukesbrave/digital-home-backend-starter into PROJECTS/DIGITAL HOME V1/Digital_home_backend_starter/.
• Resend domain verification pending: You need to verify sondraverva.com in the Resend dashboard and update RESEND_FROM_ADDRESS in .env.local.
• Next.js workspace root warning: This requires a Claude Code session to add turbopack.root to next.config.ts.
• Claude Code Setup - Agent teams not yet enabled: You need to manually add "env": { "CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS": "1" } to ~/.claude/settings.json before trying agent teams, and verify Anthropic billing first.
• Claude Code Setup - Delete duplicate skill files from module-installs/: This requires a Claude Code session to delete module-installs/youTube_search_skill.md and module-installs/youtube_pipeline_skill.md.
• Skool Admin Dashboard - Dashboard posting not yet confirmed end-to-end: This requires a Claude Code session to diagnose the Flask server issue and confirm SKOOL_EMAIL and SKOOL_PASSWORD in .env.
• Skool Admin Dashboard - Post as Sondra or as AI Admin?: This is a decision for you regarding the long-term posting strategy.
• AIOS Dashboard — Obsidian Integration:
• Custom Frames vs full custom Obsidian plugin?: This is a decision for you on how to integrate the dashboard inside Obsidian.
• INSTALL-AUTO-LAUNCH.bat not yet run: You need to right-click → Run as administrator on agentic-os-dashboard/INSTALL-AUTO-LAUNCH.bat.
• Notion Task Scheduler triggers not yet disabled: You need to disable (not delete) push_to_notion.py and notion_context.py in Task Scheduler.
• Thumbnail Generator - 10 Facebook photos unconfirmed: You need to visually confirm and properly classify the expression/pose for the 10 Facebook photos in catalog.json.
• Thumbnail Generator - search_examples.py yt-dlp path bug: This requires a Claude Code session to fix the yt-dlp path issue using shutil.which('yt-dlp').
• Thumbnail Generator - Gemini face likeness quality: You need to evaluate the face likeness quality of clean-v1.png and decide on a fallback option if insufficient.
• YouTube Project - Full YouTube workflow not yet built: This requires a Claude Code session to build the complete system, starting with competitor research.
• YouTube Project - Competitor research pending: This requires a Claude Code session to find 3 AI educator channels at ~1k subscribers, pull top thumbnails, and audit them.
• Base44 Community Site - Which Amplif-AI portal app to develop further?: This is a decision for you to pick one of the two built options.
• Base44 Community Site - Lovable MCP not yet tested: You need to test https://mcp.lovable.dev on a landing page build.
• AIOS / Skool Tools - Daily cross-post community URLs: You need to provide the Skool community slugs to finish the cross-posting setup.
• AIOS / Skool Tools - Weekend hourly posting queue: You need to provide pre-written content for the posts/queue/ folder.
• AIOS / Skool Tools - Higgsfield CLI: You need to run higgsfield auth login and define its use case.
• AIOS / Skool Tools - Skool category IDs: This requires a Claude Code session to map other category IDs.
• AIOS / Data Integrity - What is the actual current total Skool member count?: The current-data.md states 603 free members as of June 1, 2026.
• Tools & Privacy - Littlebird.ai — worth trying?: You need to evaluate the free tier of Littlebird.ai for one week.
• Tools to Evaluate - Claude Memory Compiler: You need to evaluate this tool for auto-capturing sessions and building a searchable knowledge base.
• Tools to Evaluate - Context Engineering Intro: You need to compare this planning workflow with /plan-feature before deciding on installation.
• Tools to Evaluate - Open CoDesign first run: You need to test this tool in practice, setting up the API key in Local CLI mode.
• Income / Financial - Other affiliate programs (non-Skool): You should check bank statements for unidentified income sources from other affiliate programs.
• AIOS / Claude Code - Will Claude Code see retroactive edits to daily notes?: Process clarification is needed on how Claude Code handles late edits to daily notes.
• AIOS / Claude Code - Can YouTube transcripts be auto-pulled into Obsidian/Claude Code?: No solution is implemented yet for auto-pulling transcripts from videos watched on phone/TV.
• AIOS / Claude Code - How to move OneNote content into Obsidian: This requires a plan to migrate your OneNote content.
• AIOS / Claude Code - NotebookLM notebooks and Claude Code: The best workflow for referencing NotebookLM notebooks through Claude Code is unclear.
• Business Strategy - Move everything to GHL snapshot or keep paying $97/mo?: This is a decision for you regarding your GHL setup.
• Business Strategy - Base44 affiliate issue: You need to investigate why you are not receiving affiliate money from Base44 referrals.
• Tools and Products - Hermes Agents desktop: This tool needs to be investigated to determine its value.
• Apps & Products - Viktor templates decision: You need to decide whether to package and sell, give away, or both, the 19 Lovable hero/animation templates.
• Apps & Products - Stars DFY apps 7-10: You need to plan what the remaining 4 DFY apps for Stars members will be.
• Apps & Products - Base44 public domains: This requires a Claude Code session to run save_base44_session.py + collect_base44.py to get real public URLs and a true app count.
• Apps & Products - More Base44 apps likely untracked: This requires a Claude Code session to run a script to reveal how many apps have been built since the last Notion update.
• Offers and Revenue - Courtney website pricing: You need to decide on a price for the Base44 website clone that two people want to buy.
• Business Strategy (new — 2026-05-09) - Impeccable Chrome extension wrapper → service business: You need to evaluate using Impeccable as a wrapper to offer a website analysis/redesign service.
• Git / Skills to Evaluate - https://github.com/browser-use/video-use: This tool needs to be investigated.