instructionChain (spec 3.0.0, filesystem.md): from the boundary — the project root — down to the start directory, outermost first, each directory gives the first of AGENTS.md, CLAUDE.md that is a file really inside the project; one that is not (a link out) is passed over and the walk goes on below it. Before, only the nearest file counted. The chain's files are named to the model by their path in the project; agentsBlock reads them all: none means "make one", and git is settled when any of them settles it. Conformance runner for the `instructions` area (all 9 cases pass). Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
LLeMbas CLI
A terminal coding agent and project manager: the lembas command. It talks to any LLM API
(OpenAI-compatible servers such as llama.cpp, vLLM or LM Studio, the OpenAI Responses API,
Anthropic, Gemini, Ollama), works in your project with four permission modes and a git snapshot of
every turn, remembers what it learns across sessions, and can sign in to a
LLeMbas server to use its models, library and search — and
let its web UI work the same sessions. Linux, x64 and arm64.
Install
curl -fsSL https://github.com/LLeMbas/LLeMbas-CLI/releases/latest/download/get.sh | bash
Or download a binary from Releases. Every
release's SHA256SUMS is signed with the release key, and get.sh checks the signature and the
checksum before installing anything (SECURITY.md shows how to check by hand).
First steps
cd your-project && lembas # the TUI
lembas login https://your-server # sign in to a LLeMbas server and use its models
Without a server, add a connection to ~/.config/lembas/connections.yaml and set model: in
~/.config/lembas/config.yaml.
lembas updates itself (lembas update does it now); lembas uninstall removes it, and
--purge takes settings and sessions too.
Links
- Documentation: https://llembas.eu/cli.html and https://llembas.eu/docs/
- Changelog: https://llembas.eu/changelog.html
- The server: https://github.com/LLeMbas/LLeMbas
Licence
MIT — see LICENSE and THIRD_PARTY.md.