I Built Context+ AST and Embeddings for Codebase Understanding
AST + embeddings for codebase search—but Sourcegraph Cody, Cursor, and Continue already solve this.
Sift through code for embedded authorization logic. Extract it into Policy as Code.
Tree-sitter plus LLM hybrid finds auth logic Semgrep rules miss.
Backend developers, security engineers, DevSecOps teams
Semgrep · CodeQL · OPA
https://github.com/EnforceAuth/zift
Written in Rust, so it hums through code.
Supports JS/TS/Java/Go/Python/C# now, more coming. Outputs Rego for OPA now, Cedar coming soon.
You can hook up your local agent for a `--deep` scan
AST + embeddings for codebase search—but Sourcegraph Cody, Cursor, and Continue already solve this.
Tree-sitter + agentic flow ~50x faster than raw LLM calls, but codebase visualization is crowded category.
Tree-sitter + FTS5 + MCP = tokens saved for AI agents to actually code, not search.
Tree-sitter + LLM codebase mapping, but Cursor, Continue, Sourcegraph already do this.
Knowledge graph RAG for code when Cursor and Cody already own this space.
Graph-augmented RAG beats pure vector search, but codebase Q&A tools are crowded.