Astrail turns the Instagram Reels you save into a routed trip on a live 3D map, and you build and edit the whole thing by talking to ChatGPT. The WebMCP tools run inside the page, not on a server.
Aggregate36
Leverage9
Execution9
Impact9
Creativity9
Each criterion 1–10, equally weighted; aggregate is their sum. Ranking is the pipeline's consolidated output.
Origin, access model, eligibility, and substitution are reviewer diagnostics, not judging criteria. Authentication requirements are not penalized.
02 The two blind reviews
Two independent reviewers scored this project blind, from a sanitized evidence packet.
Scores are shown separately so the reasoning stays inspectable. A withheld score means the reviewers differed by more than two points.
Reviewer A (round 1)
confidence 62%
Leverage
8 /10
Structured in-page tools let the agent operate on saved reels, itinerary state, and map edits with shared state rather than fragile multi-tab UI driving; the same outcome is possible manually but substantially less reliably.
Evidence cited
Pitch says WebMCP tools run inside the page and save pasted Reel links/start extraction.
About text describes visible add_place calls, recalled preferences, verification, routing, and reporting missing forecast rather than fabricating it.
Execution
7 /10
The described central workflow is coherent and appears substantially implemented, but the packet has no transcript and image review was unavailable, so end-to-end completion is claimed rather than observed.
Evidence cited
Demo is marked alive and public repository exists.
About text describes staged extraction, verification, approval, routing, and adding a user-owned stop.
Impact
8 /10
Targets a concrete travel-planning pain point for people with inspiration saved in social media and converts it into an actionable itinerary.
Evidence cited
About text identifies the gap between saved Reels and usable trip plans.
Itinerary includes routes, geography-based days, and weather/verification handling.
Creativity
8 /10
The social-video-to-live-3D-trip-planning concept is a fresh, memorable synthesis with a meaningful conversational editing model.
Evidence cited
Combines Instagram Reel inspiration, multi-agent extraction, verification, and live 3D mapping.
Agent-added stops are explicitly distinguished from Reel-derived places.
Reviewer B (round 2)
confidence 77%
Leverage
8 /10
The shared page state lets ChatGPT save reels, build/edit an itinerary, and update the 3D map through structured tools; a general UI agent could do this but with substantial friction and less precision.
Evidence cited
Description states WebMCP tools run inside the page and shared state is visible beside ChatGPT.
Frame collage shows chat, saved reels, place cards, itinerary, map, approval, and route-redraw states.
Natural-language request to add Tokyo Disneyland is depicted as an itinerary/map edit.
Execution
7 /10
The packet shows a coherent progression from saved content to mapped itinerary and edit, but static collages and small text leave external verification and exact tool execution unproven.
Evidence cited
Frame sheets show consistent chat, itinerary, place-card, and Mapbox-like 3D map surfaces.
Visible captions describe approval, verification/drop rules, and route redraw after adding a stop.
Devpost screenshot is a paused video thumbnail, not runtime proof.
Impact
8 /10
Travelers with large collections of saved inspiration have a specific planning problem, and converting those saves into a usable, routed trip is a credible high-value solution.
Evidence cited
Description identifies the gap between saved Reels and a real itinerary.
Workflow includes verified places, preferences, weather, routes, and restaurants.
The result is a day-by-day mapped plan that can be edited conversationally.
Creativity
8 /10
The product is a memorable synthesis of social-media inspiration, conversational planning, multi-agent verification, and live 3D itinerary editing.
Evidence cited
Saved Instagram Reels become verified geographic stops rather than merely bookmarks.
Conversational edits redraw the mapped route.
Missing forecasts are reported rather than fabricated, per the description.
03 Review highlights
Standouts across reviewers
Strong bridge from saved social content to an actual route.
Strong before/after route-edit story.
Clear conversion of unstructured saved media into a structured trip.
Verification and explicit approval are integrated into the experience.
Red flags
No submitted video or transcript; visual claims could not be independently checked.
Static collages cannot verify live imports, APIs, coordinates, or persistence.
Some visual captions and labels are small or inconsistent in naming.
No transcript was supplied.
04 Evidence
What each artifact proves is labeled on the artifact itself. A Devpost page capture is packaging evidence, not proof the product runs;
video frames are evidence from the submitted demo, not live verification.
Contact sheets from the ?s video the team submitted. This is what reviewers were shown;
it demonstrates the product in motion but is not independent verification.
· watch the original
EX-V1 Sheet 1 of 3 · reported video evidenceEX-V2 Sheet 2 of 3 · reported video evidenceEX-V3 Sheet 3 of 3 · reported video evidence
