Structure

Layers are inferred from where files sit in the tree, not from reading the code. A project that names its directories unconventionally will read oddly here — open the file browser to check anything the diagram implies.

Languages

Share of indexed source by file size. Binary and vendored files are excluded.

Dependencies

Declared in the repository’s manifests at the indexed commit. A declared package is not proof it is used, and runtime dependencies are listed first.

Feature verification

• AI persona generation via ClaudeVerified

  Generate a simulated prospect persona based on learning goals, company-specific data, and industry research using LLMs

  Claimed on readmehigh confidence

  • backend/internal/integrations/claude.go:28 — GeneratePersona builds a prompt from role/company/difficulty/notes and calls the Anthropic Messages API, parsing the JSON persona from the response
  • backend/internal/services/persona_service.go:23 — PersonaService.GeneratePersona orchestrates the Claude call and returns a persona to the handler
  • backend/internal/handlers/persona.go:24 — POST /api/persona/generate endpoint wired to the service
• Go backend for API layerVerified

  Built with Go for the backend

  Claimed on Devposthigh confidence

  • backend/go.mod — Go module defining the backend
  • backend/cmd/server/main.go:18 — Gin HTTP server exposing /health and /api/persona/generate
• Learning-goal and difficulty customization for the pitch scenarioVerified

  Simulation is catered based on prospective company, sales rep's learning goals, and difficulty

  Claimed on readmehigh confidence

  • frontend/app/page.tsx:16 — Intake form collects prospectRole, companyName, callType, difficulty, and notes
  • backend/internal/integrations/claude.go:99 — buildPersonaPrompt feeds role, company, difficulty, callType, and notes directly into the Claude prompt
• Next.js/TypeScript/TailwindCSS/React front-endVerified

  Built with NextJS, TypeScript, TailwindCSS, ReactJS for the front-end

  Claimed on Devposthigh confidence

  • frontend/package.json — Lists next, react, tailwindcss, typescript as dependencies
  • frontend/app/page.tsx:1 — App Router page using 'use client' React components
• Video call UI (camera, mic controls) for the practice sessionVerified

  Implied by 'mock sales pitch calls' and 'speaks to you in real time' framing of the simulation experience

  Claimed on readmehigh confidence

  • frontend/app/simulation/page.tsx:67 — getUserMedia captures webcam/mic, with mute/video toggle buttons and a call timer wired to real state
• Upload of previous client calls for analysisCode-supported

  Implied by 'company-specific data' input and the 'Previous client calls' upload field on the intake form

  Claimed on readmehigh confidence

  • frontend/components/FileUpload.tsx:13 — FileUpload component lets a user select an mp3/mp4/mov file and calls onChange(file), but the file is only kept in local component state
  • frontend/app/page.tsx:93 — Home page renders FileUpload with setUploadedFile as onChange, but uploadedFile is never sent to the backend or referenced elsewhere in handleSubmit
• Voice greeting via ElevenLabsCode-supported

  The AI prospect speaks to you (voice synthesis via ElevenLabs)

  Claimed on readmemedium confidence

  • backend/internal/integrations/elevenlabs.go:28 — GenerateGreeting builds a short greeting line and calls ElevenLabs text-to-speech, returning a base64 audio data URI
  • frontend/app/simulation/page.tsx:43 — Simulation page plays the one-shot greeting audio once on load via an <audio> element
• AI agents acting autonomously on company data (Rox-like agent behavior)Claimed only

  AI agents (similar to what Rox offers) act autonomously on a company's messy, real-world data

  Claimed on Devposthigh confidence
• Analysis of company's real-world sales data / industry researchClaimed only

  Analyzes a company's messy, real-world sales data and industry research to build the persona

  Claimed on readmehigh confidence
• Post-call coaching feedback with scores, strengths, weaknesses, suggestionsClaimed only

  Implied product flow: practice call followed by structured feedback ("learn, practice, and apply"); enableCoaching toggle exists on intake form

  Claimed on readmehigh confidence
• Real-time two-way voice conversation with AI prospectClaimed only

  This AI prospect speaks to you in real time (implying an ongoing back-and-forth pitch call, not a single greeting)

  Claimed on readmehigh confidence

An AI agent derived these features from the project’s Devpost page and readme, then searched the code for each one. Verified features are backed by cited code; claimed-only features had no supporting code, which is not by itself proof a feature is missing.