mex alternatives
Curated alternatives to mex — and why you'd switch.
Graft
Context layer for large codebases: a graph of plain-English markdown nodes — no embeddings, no index — agents read like any repo file. Claude Code hooks + MCP; 42% fewer tokens in its bench.
Why switchBoth build a Markdown context layer over a large codebase with no embeddings, served through agent hooks and MCP. Graft keeps it to a graph of prose nodes agents read like files; mex puts a Tree-sitter symbol graph underneath so claims can be grounded and drift detected.
Full comparison →Trellis
Spec, task and memory layer that lives in your repo: .trellis/ holds conventions, PRDs and journals, auto-injected each session — one workflow across 20 coding-agent platforms.
Why switchTwo in-repo memory conventions for coding agents. Trellis is spec- and task-shaped — conventions, PRDs and journals in `.trellis/` across 20 platforms; mex is knowledge-shaped, with a router, a code graph and drift checks. Process memory versus codebase memory.
Full comparison →brain.md
File-based durable memory for coding agents: brain-setup scaffolds a BRAIN.md protocol + brain/ directory of decisions, requirements and constraints — plain Markdown in your repo, written via CLI.
Why switchBoth keep durable agent memory as plain Markdown in the repo. brain.md is a protocol and CLI you can read in a minute; mex is the heavier system — graph, routing, grounding, sync — for when a project outgrows a single file.
Full comparison →