01 · Roasts
Speed Runner (Not the Good Kind)
house_prediction was committed in 3 minutes, supergent in 2 minutes, and extension in 3 hours. Vicky, these aren't repos — they're snapshots of a train of thought.
85% TypeScript, 0% Tests
Your codebase is 85% TypeScript — a language literally designed to catch bugs — yet only 2 of 11 repos have any tests at all, and both sets are placeholder-grade. The type checker is doing all the heavy lifting while your test suite sleeps.
Idea Factory, Shipping Drought
Seven distinct AI agent projects (runway_insta_agent, astron-browser, superclaw, superogent, supergent, supermail, extension) and zero of them have crossed 2 stars. At this point 'super' is a prefix, not a promise.
The Drop-Off Is Real
Heatmap weeks 1–37 look genuinely active. Weeks 38–52? A graveyard of zeros. Your GitHub is either seasonal or you've already moved on to the next idea before the last one compiled.
One CI to Rule Them All
Out of 30 public repos, exactly ONE has a CI pipeline — your portfolio website. Everything else ships untested and unvalidated. The irony of deploying a portfolio that proves you know CI exists is not lost.
Built using
Zoral
Shadows one worker for a week, then takes over their job with zero extra setup. Behaves exactly like the original.
zoral.ai
02 · Category breakdown
- Impact25% weight68C
- Consistency20% weight45D
- Quality20% weight57D
- Depth15% weight37F
- Breadth10% weight55D
- Community10% weight25F
03 · Stats
365-day commit heatmap
280 active days
Language distribution
- TypeScript85%
- Jupyter Notebook10%
- Python2%
- CSS1%
- MDX1%
- JavaScript1%
04 · Numbers
Owned repos
non-fork
24
Commits
last 12 months
170
Followers
5
Joined GitHub
May 2021
05 · Top repos
manideep1428 /
astron-browser
Terminal UI wrapper for browser-use agent with multi-model LLM support. Typed TypeScript + React/Ink, structured src/ & python/, config management. Created Feb 2026, 6 commits in ~7 days—early-stage but functional project.
manideep1428 /
superclaw
Early-stage TypeScript web app providing browser UI for OpenClaw extensions; documented with QUICK_START.md, ARCHITECTURE.md, STATUS.md; no tests/CI yet; typed but minimal commit history (2 of last 30 days).
manideep1428 /
runway_insta_agent
Early-stage Instagram content automation agent combining RunwayML, Composio, and Claude/GPT APIs with React Flow visual editor. Typed TypeScript, documented README, but minimal tests/CI and nascent git history.
manideep1428 /
house_prediction
One-day house price ML tutorial demonstrating end-to-end sklearn workflow (data generation, EDA, Linear Regression, Random Forest with GridSearchCV) with structured src/ layout, tests, and synthetic visualizations—educational but with synthetic data and no domain application.
manideep1428 /
manideep1428.github.io
Personal portfolio website built with Next.js and TypeScript. Well-structured React components, typed code, and CI/CD pipeline. Minimal production impact as an individual portfolio project.
manideep1428 /
superogent
TypeScript Next.js project for AI-powered code generation in sandboxes. Has README, typed language, structured layout (app/, components/, ai/), and functional integration with Vercel APIs. However, appears to be a recent scaffold (created 2026-05-01, only 1 commit), lacks tests, CI, production adoption signals.
manideep1428 /
supermail
Early-stage OpenEnv RL challenge submission: deterministic email triage environment with typed models, structured tasks, and FastAPI server. Minimal maturity (2 stars, <30 commits in 1 day), no tests or CI.
manideep1428 /
extension
Early-stage Chrome extension for fact-checking Instagram/X posts using OpenAI. Typed TypeScript code, modular structure, but lacks documentation, tests, CI, and has only 6 commits in 3 hours.
manideep1428 /
bhai
Fresh VS Code extension (2 stars, 2 commits in 2 days) that plays audio on keypresses. TypeScript + tests + MIT license, but minimal scope and experimental trajectory.
manideep1428 /
insta
Personal Instagram auto-uploader tool built in TypeScript with Express server. Polls BunnyCDN for videos and publishes them to Instagram via Graph API on a 6-hour schedule. Minimal documentation, no tests/CI, lightweight codebase (~24 KB).
manideep1428 /
supergent
Early-stage Next.js project scaffolding an AI-powered code generation platform using Vercel Sandbox and OpenAI. Minimal commits (3 in 2 minutes), no tests/CI, incomplete implementation with placeholder UI components and untested tool integration.
06 · Timeline
- May 13, 2021Joined GitHub
- Feb 28, 2026Created astron-browser — Browser Automation tool (Claude code for the browser)
- Mar 19, 2026Created manideep1428.github.io
- Mar 22, 2026Created superclaw
- Apr 4, 2026Created extension
- Apr 8, 2026Created supermail
- Apr 13, 2026Created insta
- Apr 14, 2026Created bhai
- Apr 17, 2026Created house_prediction
- Apr 24, 2026Created supergent
- May 1, 2026Created superogent
- May 11, 2026Created runway_insta_agent
- May 15, 2026Most recent push to runway_insta_agent
07 · Compare
08 · Rubric
How this score was produced
Overall = Σ (category × weight) + gentle top-end curve
Tier thresholds
▸ How the pipeline works
- 01Scrape.Pull every non-fork repo pushed in the last 90 days, plus your contribution calendar, followers, and language byte counts — straight from GitHub's REST & GraphQL APIs.
- 02Triage.A small model reads every repo's file tree + README and picks the 20 files per repo that actually reveal how you code.
- 03Grade each repo. All repos run in parallel through a fast scoring model that reads the picked files and rates each one independently on Impact, Quality, and Depth — with evidence citations.
- 04Aggregate. A larger reasoning model combines the per-repo scores with server-computed stats (heatmap, commit cadence, language entropy, follower count) to produce the 6-dimension profile score + roasts.
- 05Correct.Deterministic server-side checks enforce anchor-scale floors (e.g. a profile with 2,000+ public commits can't score 30 Consistency) and recompute the final verdict.
~90 seconds per profile, ~$0.25 in compute. Total of ~240 files read across your top-12 repos. One rating per GitHub account per day.
▸ Data sources & caveats
- Heatmap & commit totals: GitHub GraphQL
contributionsCollection— covers the last 365 days, includes private repos when the user has opted in (default). - Language %: byte totals across the top 30 owned non-fork repos.
- Curve: a small upward nudge centered on raw score ≈ 70, capping at 100. Prevents specialists from being unfairly penalised for narrow breadth.
- Anchor corrections: when server-measured signals (e.g. privateWorkLikely, multiRepoVolume, follower count) mandate a minimum category score, the aggregation step enforces it. These are signal-conditional, not identity-based floors.