Zift – find authorization logic in your code
Tree-sitter plus LLM hybrid finds auth logic Semgrep rules miss.
Mapstr analyzes any software project and generates a complete codebase map in seconds. Outputs a natural-language summary, a Mermaid dependency graph, and structured JSON — powered by Claude, OpenAI, Gemini, or any local model.
Tree-sitter + LLM codebase mapping, but Cursor, Continue, Sourcegraph already do this.
Engineers onboarding to new projects, architects documenting systems
Cursor AI · Continue IDE · Sourcegraph Cody
Pre-flight API checks (Groq/OpenAI/etc), cost tracking, cache. go install github.com/BATAHA22/mapstr@latest
Tired of reading docs? Map it. v1.4.0 out now!
Tree-sitter plus LLM hybrid finds auth logic Semgrep rules miss.
Tree-sitter + agentic flow ~50x faster than raw LLM calls, but codebase visualization is crowded category.
Knowledge graph RAG for code when Cursor and Cody already own this space.
Deterministic cross-file impact analysis that catches breaking changes LLMs and linters miss.
AST + embeddings for codebase search—but Sourcegraph Cody, Cursor, and Continue already solve this.
Tree-sitter + FTS5 + MCP = tokens saved for AI agents to actually code, not search.