AKF
AI native file format for trust metadata and provenance tracking
Visit akf.dev ↗What is AKF?
AKF is a file format that embeds trust metadata, provenance information, and compliance data into files touched by AI agents. It works like EXIF for AI—stamping files with verification state, confidence scores, and source information so subsequent agents or humans know what has already been verified. A stamp costs roughly 15 tokens, while re-verifying the same work costs 15,000 tokens.
The system supports 20+ file formats including code, documents, images, and media. Agents can install AKF hooks into their workflows (git, Claude Code, rules files) to automatically stamp verified work and check existing stamps before rebuilding on them. The maker indicates it was built with Claude Code and compatible with tools like Cursor and Windsurf.
Momentum
Daily activity over the last 30 days: upvotes, unique visits, new GitHub stars and Hacker News points.
How do we know it's vibe coded?
vibetrend discovered AKF automatically. Evidence: GitHub HMAKT99/AKF; repo has .cursorrules, .windsurfrules, AGENTS.md. View source.
Similar vibe-coded apps
Turn your codebase into a queryable knowledge graph for AI coding assistants
Operating layer for AI coding agents with skills, memory, planning and security
Self-hosted AI agent with persistent memory and multi-platform messaging
Open-source AI platform for short-form video and animated series creation
Open-source canvas workbench for AI-powered image and video creation
Community marketplace for DeepSeek Harness Web themes, pets, plugins
Remote browser infrastructure and agents for AI automation tasks