Media
4 items
Description
Solo Associative Remote Viewing trials. Your tasker, judge, and feedback — voice-prompted, AI-judged, blind-protected. SoloARV lets one person run a complete Associative Remote Viewing trial on their own. The extension plays Tasker, Judge, and Feedback so the viewer stays blind to both target images until the experiment is finished. WHAT IS ASSOCIATIVE REMOTE VIEWING? Associative Remote Viewing (ARV) is a research protocol where a viewer attempts to perceive a future image they will be shown — as a way of indirectly predicting a binary outcome. Two visually distinct photographs are pre-assigned to the two possible outcomes (Up/Down, Yes/No, Team A/Team B, and so on). The viewer sketches their first impression without seeing either image, and a judge later compares the sketch to the two photographs to pick the closer match. That match is the predicted outcome. The traditional protocol requires a separate Tasker (who frames the question), a Viewer (who sketches), and a Judge (who scores the match). Solo practice is hard because the viewer cannot stay blind to the targets without a second participant. SoloARV solves that by automating every role except the viewer. HOW IT WORKS — FIVE CLEAR PHASES 1. SETUP Enter two outcome labels, describe the event you're predicting, and choose your feedback time. SoloARV randomly picks two distinct photos from its built-in pool of 800 curated images, randomly assigns each to one of your outcomes, and encrypts the mapping with AES-GCM-128. You never see either image during Setup — the viewer stays fully blind until feedback. 2. TASKING When you're ready, press play and SoloARV reads a calm, measured tasking prompt aloud using your operating system's text-to-speech voices. Then sketch your first impression on the in-app canvas — or upload a photo of a hand-drawn sketch or a multi-page transcript instead — and add written notes: shapes, textures, colors, scale, mood, anything. Submit whenever you're ready. 3. JUDGING Your sketch, your notes, and both target photographs are sent to Anthropic's Claude vision model as an impartial judge. The photographs are labeled neutrally as "Image A" and "Image B" so no outcome label can leak into the model's reasoning. The judge returns a winner (A, B, or NA), a 1-to-3 correspondence score, and written reasoning citing the specific visual or descriptive details that matched. 4. RESULTS SoloARV maps the judge's winner back to one of your outcomes and shows a direction code (e.g., "U-2" = Up with Medium correspondence), a plain-language recommendation, and the model's reasoning. The two target images remain hidden. 5. FEEDBACK When the real-world event resolves, you tell SoloARV which outcome actually happened. SoloARV reveals only the winning image — the losing image is never shown — and produces a downloadable JSON record of the trial. FIRST-RUN SETUP — YOU PROVIDE YOUR OWN ANTHROPIC API KEY SoloARV uses Anthropic's Claude AI as the impartial judge that compares your sketch to the two target images. To use the judging feature you need your own Anthropic API key. Setup takes about 5 minutes: 1. Sign up at https://console.anthropic.com (free account, paid usage). 2. Add a payment method and load credits — most users start with $5–$10. The API has no free tier, but typical judging cost is roughly $0.05 to $0.15 per trial, so $5 covers 30 to 100 trials. 3. In the Anthropic console, go to Settings → API Keys → Create Key. Give it any name (e.g., "SoloARV"). Copy the key — it starts with "sk-ant-api03-…" and is shown only once. 4. In Chrome, right-click the SoloARV toolbar icon and choose Options (or open Settings from the popup). 5. Paste the key into the "Anthropic API key" field, click Save, then click Test connection. A green "Connection OK" confirms it works. Your key is stored only on this device, in chrome.storage.local. It is sent only to api.anthropic.com when you initiate a judging request. The SoloARV authors never see it. If you don't have an Anthropic account yet, SoloARV will not be able to judge sketches. Setup and tasking still work offline; only the judging step requires the key. FREE FOR THE FIRST 10 TRIALS SoloARV is free for the first 10 trials. After that, continued use requires SoloARV Pro, a subscription ($9.99/month or $99/year) available at https://www.soloarv.com/pricing. Your saved trial history and statistics remain available even without Pro. To activate Pro, paste the license key you receive after purchase into Settings → Pro license. PRIVACY — LOCAL-FIRST BY DEFAULT Trial state, settings, your API key, target images, and the encrypted mapping all live on your computer. There is exactly one outbound network call: the judging request to api.anthropic.com. What does NOT happen: - No analytics - No telemetry - No tracking pixels - No third-party scripts - No remote code - No cookies - No accounts You can wipe everything in one click from Settings → Clear all stored trial data (with double confirmation). The full privacy policy is at https://www.soloarv.com/privacy.html. WHAT SOLOARV IS NOT SoloARV is a research tool and a structured decision-making framework. It is NOT financial advice. It does NOT predict the future with any guaranteed accuracy. It does NOT replace your own due diligence in trading, betting, or any other domain. ARV correspondence scores are observational and should be treated as such. REQUIREMENTS - Chrome 120 or newer - An Anthropic API key (sign up at console.anthropic.com) - A working text-to-speech voice on your operating system (built into macOS, Windows, and ChromeOS) - An internet connection at the moment of judging (everything else works offline) DRAWING TIPS You have two ways to submit your viewing: sketch directly on the in-app canvas (brush, eraser, color, undo, clear), or upload a photo of a hand-drawn sketch or a multi-page transcript instead. If you upload, use a dark pen, marker, or Sharpie on white paper — the AI judge needs to clearly see the lines. Pencil and faint marks often fade in the photo and weaken the judging. FREQUENTLY ASKED Q: Why bring my own API key instead of paying you? A: To keep SoloARV free, fast, and free of any intermediary server. Your data goes straight from your browser to Anthropic — it never touches a SoloARV server, because there isn't one. Q: Can SoloARV peek at the target images? A: The extension stores them locally so it can send them to the judge and reveal the correct one later. The image-to-outcome mapping is encrypted with AES-GCM-128 using a per-trial key, so a casual look at chrome.storage.local will not tell you which is which. Q: Does it work offline? A: Setup, the voice prompt, drawing, taking notes, saving records, and reviewing past results all work offline. Only the judging step (Phase 3) needs internet. Q: Which Claude model is used? A: claude-sonnet-4-5 by default. You can edit one line in background.js to swap to claude-haiku-4-5-20251001 (cheaper, faster) or claude-opus-4-6 (more capable, more expensive). The README inside the extension explains the swap. Q: What if my judging request fails? A: SoloARV shows a clear error and a Retry button. Your sketch and notes are saved locally, so retrying does not require redrawing. Q: Can I run two trials at once? A: One trial at a time, by design. Start a new trial only after the current one reaches Feedback. Q: Does SoloARV claim to predict the future? A: No. It is a structured framework for running ARV experiments. Treat the score as an observation, not a guarantee. SUPPORT Email support@soloarv.com for questions, bug reports, or feedback. The full source is fully inspectable — right-click the extension and choose "Inspect popup" to see exactly what is running.