THOR runs on your own machine — the client starts it, so there is no endpoint to ping. Last commit 20 Oct 2026.
Local memory for coding agents; a rule with a proof can refuse. Full install with hooks: the repo.
What the code does
We read the source, 5 d ago · rules 995398eba69f
What this server is able to do. For an MCP server this is often the job itself — a terminal server runs commands because that is what it is for. Listed so you know what you are plugging in, not as an accusation.
# The shared secret arrives as THOR_SYNC_TOKEN (env, never a CLI flag: a # secret on the command line ends up in `ps` and in shell history). The # receiver refuses to start without it rather than listening unauthenticated. ENV THOR_SYNC_TOKEN=""
Is this your server and something here is wrong? Tell us — corrections are free and do not require a plan.
This code can reach further than it looks
We found places where it runs commands, builds paths or queries from values it is given. None of that is a flaw by itself — it becomes one when the code changes, and code changes quietly between releases. We re-read it on every one.
Alternatives to THOR
same job, measured the same wayPersistent project memory for AI coding agents: one compact digest instead of re-reading the repo.
Persistent local memory for MCP-compatible AI coding agents, stored per repository in SQLite.
Local-first memory for AI coding agents. Memory + CodeGraph + Wiki in one SQLite file.
Local-first memory for AI coding agents: shell history with exit codes, git diffs, docs, MCP.
Local codebase memory and repository code search for Codex, Claude Code, and MCP coding agents.
Session memory for coding agents: local-first recall, drift detection, on-device inference.
Shared, git-tracked working memory for AI agents on the same codebase.
Project memory across sessions for AI agents. A queryable decision log in the repo.
THOR — questions
Answers built from our own checks of this server.