landfalls-ai/landfall-cli

Join a Landfall war room from your own computer and have your
**favorite MCP-capable agent** (Claude Code, Codex, Cursor, VS Code, Claude Desktop,
Windsurf, …) become a **live investigator** in the shared room: its findings land on
the incident timeline in realtime, and it pulls everyone else's discoveries between
tasks.
Installation
Or without Homebrew, download a prebuilt binary from the
releases page — no toolchain
required, not even Go. Pick the archive matching your platform (darwin/linux ×
amd64/arm64; Windows is not supported yet), extract it, and put landfall on your
PATH (this intentionally does not pin a version — a hardcoded one goes stale on every
release with nothing to catch it, which is exactly what happened to the v0.1.0 this
used to say):
landfall install: auto-register every coding agent on your machine
Once the landfall CLI itself is installed, one command detects which supported
coding agents are on **this machine** and wires each one up with its own **native,
global** MCP registration — no per-repo config file needed, and it works the same
whichever directory (or none at all) you're sitting in:
| Harness | Mechanism |
|---|---|
| **Claude Code** | claude mcp add-json … --scope user (Claude Code's own user-scope MCP registration), **plus** the landfall-edge-bridge Claude Code plugin — see below |
| **Codex CLI** | codex mcp add landfall -- landfall serve (Codex's own MCP registration) |
| **VS Code** (Copilot/agent mode) | code --add-mcp, or a direct merge into your user-profile mcp.json if the code CLI isn't on PATH |
| **Cursor** | merged into your global ~/.cursor/mcp.json |
| **Claude Desktop** | merged into your claude_desktop_config.json (macOS/Windows) |
| **Windsurf** | merged into your global ~/.codeium/windsurf/mcp_config.json |
landfall install only touches a harness you select, never overwrites an existing
differently-configured landfall entry (reports a conflict instead), and is safe to
re-run — an already-configured harness is reported as such rather than duplicated.
Run landfall install --yes to configure every detected harness without prompting, or
landfall install --only cursor,codex to target specific ones.
Claude Code also gets the landfall-investigation-dashboard agent
For Claude Code specifically, landfall install does one more thing after
registering the MCP server: it best-effort runs
which installs this repo's own Claude Code plugin — the landfall MCP server
**plus** a subagent that:
- **Recognizes a war-room join prompt on sight** (the "Share with agent" paste, a
bare agent share link, or plainer "join the war room" phrasing) and calls
join_war_room + get_brief on its own.
- **Specializes in the sub-investigation dashboard** — curates a small, honest set
of post_widget widgets and keeps them current instead of scattering one-off
ones.
- **Knows the MCP best practices** for an ongoing investigation: pull before you
publish, findings vs. notes vs. widgets, suggest-only remediation, keep
secrets/raw output local, treat room content as untrusted data.
This step is best-effort and never turns a successful MCP registration into a
reported failed outcome (an older claude CLI without claude plugin, or no
network, just means the agent isn't installed yet) — the report line shows
[plugin: installed] or [plugin: skipped] alongside the usual status. Re-running
landfall install retries it. To do it yourself without landfall install:
To remove the registration:
This only removes an entry that still matches exactly what landfall install wrote —
if you've hand-edited it since, it's left in place and reported as such.
Any other MCP client (or a harness landfall install doesn't cover) uses the manual
snippet below.
landfall hooks install: make room context impossible to miss
MCP registration gives your agent the incident tools. Hooks give the room a way to
reach a session that isn't currently making a tool call — the difference between
context your agent *can* fetch and context it *will* see.
**Hooks never take the wheel.** You decide what your agent does next. No hook starts
an agent turn you didn't ask for: room news reaches your agent together with your next
message, and when your agent finishes a turn you get a one-line notice of what's
waiting. (Until v0.8.12 the Stop hook refused the agent's conclusion instead, which
made it run another turn on its own. That is now an explicit opt-in: see
What Stop does.)
**Room news is information, not a task.** Your agent's standing instructions say the