Repositories
fatihcvs/tamper-evident-verification
Tamper Signal: signed receipts for vibe-coded data pipelines. Proves nobody changed your data, and shows the exact link if they did.
fatihcvs/libredb-studio
One browser tab for PostgreSQL, MySQL, Oracle, SQL Server, MongoDB, Redis, SQLite, Couchbase, ClickHouse, Druid, DuckDB, Turso and more. An open-source SQL IDE with SSO, audit trail and AI-assisted queries MIT licensed, with nothing held back behind an enterprise wall.
fatihcvs/groundhog
Replay your git history as a regression test for coding agents. Your commits are already labeled eval data.
fatihcvs/Chronicle
fatihcvs/seedgraph
Seed your SQLAlchemy models as a referentially-consistent graph — one call, shared parents, verified links.
fatihcvs/capes
Superpowers for any AI: a free, self-hosted MCP server (Gmail, Discord, X) with a dashboard. Deploy to your own Vercel in minutes.
fatihcvs/backup-coverage
Read-only Compose and restic backup scope checks with explicit evidence and inventory drift
fatihcvs/blazecrawl
Security-first, self-hostable web extraction for developers and AI systems.
fatihcvs/AyorGraph
Stateful agent orchestration framework for building reliable, tool-using AI agents with routing, memory, retries, checkpoints, and human approval.
fatihcvs/judge-reliability-kit
Diagnose why an LLM judge panel disagrees: fix the rubric, or fix the task.
fatihcvs/bitrot-demo
One flipped bit, five image formats. Demo for a dev.to post.
fatihcvs/thonny-ai
MCP server + Thonny plugin: let Claude Code or any AI agent program Raspberry Pi Pico & ESP32 in MicroPython. Visible in the shell, approvable, auto-reconnect.
fatihcvs/a2a-drift
Detect A2A (Agent2Agent) protocol compliance drift — agent cards, endpoints, spec versions, JSON-RPC conformance
fatihcvs/mutiny
Behavioral fuzz-testing engine for AI agents — policy violations, tool-call verification, regression tests. Adapter #1: OpenAI Agents SDK.
fatihcvs/batten
Engineering infrastructure-as-code. One YAML manifest wires CI, hooks, agent instructions, MemPalace, gstack, CodeGuard, and more across 7 languages and 4 AI agents.
fatihcvs/jellyfin-media-organizer
Jellyfin Media Organizer (JMO) — safe, deterministic tooling for organizing media into Jellyfin-friendly layouts.
fatihcvs/SpecSeal
Specs, sealed — an implement/review agent chain with hook enforcement, an evidence ledger with drift detection, and a tool-agnostic review handoff protocol.
fatihcvs/living-scene
Your GitHub profile README, synced to your actual weather. Animated SVG sky + seasonal garden, one Python file, zero dependencies.
fatihcvs/OpenDesigner
Open-source AI design system builder. Load it into Claude, ChatGPT, Codex or Cursor: it interviews you, recommends sourced defaults and writes DTCG design tokens, CSS, Tailwind, Figma variables, DESIGN.md and a decision log, with WCAG 2.2 checks. Built on 352 cited Decision Cards and 25 benchmarked design systems.
fatihcvs/MeshTerm
A full-featured TUI MeshCore client for your terminal
fatihcvs/claude-security-skills
Eight security skills for Claude Code with no dependencies: standard library only, tested, and usable as a CI step. Secret scanning, Python SAST, prompt-injection testing, plus HTTP, JWT, Dockerfile, CORS and dependency auditing.
fatihcvs/pulseair
Wearable air-quality risk forecasting with rigor-first evaluation — persistence floors, rolling-origin CV, GAN/SMOTE ablation, conformal prediction. Ships pulsebench, a standalone toolkit for evaluating imbalanced time-series models.
fatihcvs/E.D.I.T.H
Your Personal And Free AI Agent
fatihcvs/recusal
Deterministic governance for Claude and MCP tool calls. Pin approved capabilities, detect drift, and refuse unsafe or unapproved actions before execution. No model in the decision path.
fatihcvs/QuantEcon.py
A community based Python library for quantitative economics
fatihcvs/qr-damage
How much of a QR code can you destroy before it stops scanning? Measured, not folklore.
fatihcvs/model_router_python
Route every LLM call to the cheapest model that can actually do the job. Filters by context window, output limit and cost budget using live prices, then lets Jev pick the best model across OpenAI, Anthropic, Google and more. Zero dependencies.
fatihcvs/autosubmit
fatihcvs/cses-kit
Unofficial CSES toolkit: fetch problems, test locally, submit from the terminal.