Add lcode acp: lcode inside editors over the Agent Client Protocol

README.md

@@ -68,6 +68,8 @@ through [Ollama](https://ollama.com), and when it needs current information it c
 - **Git workflow.** `/commit` writes messages in your repository's style, `/review` checks the
   changes with `file:line` findings, `/pr` opens the pull request, each after you approve it, and
   `lcode --worktree` runs parallel sessions on one repository.
+- **In your editor.** Zed, JetBrains IDEs, Neovim and other editors that speak the Agent Client
+  Protocol run lcode with `lcode acp`, with diffs and approvals in the editor.
 - **Automation.** `lcode -p --output json` for scripts and CI, a Python API (`from lcode import
   Session`), and a GitHub Action that answers `@lcode` in issues and pull requests on your own
   self-hosted runner.

docs/editors.md

@@ -0,0 +1,85 @@
+# Editors
+
+lcode also works inside editors that support the
+[Agent Client Protocol](https://agentclientprotocol.com) (ACP), such as Zed, JetBrains IDEs,
+Neovim and Emacs. The editor starts `lcode acp` and shows lcode's work in its own UI: answers,
+tool calls, a diff for each edit, and permission questions.
+
+## Zed
+
+Add lcode to Zed's `settings.json`:
+
+```json
+{
+  "agent_servers": {
+    "lcode": {
+      "type": "custom",
+      "command": "lcode",
+      "args": ["acp"],
+      "env": {}
+    }
+  }
+}
+```
+
+Then open the Agent Panel and start a new thread with **lcode** from the new-thread menu.
+
+- **Zed can't start lcode:** use the full path in `command`, from `which lcode` (usually
+  `~/.local/bin/lcode`). An editor started from the desktop may not see your shell's `PATH`.
+- **Something goes wrong:** run **dev: open acp logs** in Zed's command palette to see the
+  messages lcode and Zed exchange.
+
+## JetBrains IDEs
+
+Add lcode to `~/.jetbrains/acp.json`, with the full path from `which lcode`:
+
+```json
+{
+  "agent_servers": {
+    "lcode": {
+      "command": "/home/you/.local/bin/lcode",
+      "args": ["acp"]
+    }
+  }
+}
+```
+
+Then pick lcode in the AI Chat tool window.
+
+## Other editors
+
+Any [ACP client](https://agentclientprotocol.com/overview/clients) works the same way: tell it
+to run `lcode acp`. That includes Neovim (CodeCompanion, avante.nvim), Emacs (agent-shell),
+VS Code extensions and more.
+
+## What the editor shows
+
+| In the editor | From lcode |
+|---|---|
+| The conversation | lcode's answers and its thinking, as they stream |
+| Tool calls | Each tool call, with its kind (read, search, edit, run, fetch) and the files it touches |
+| Edits | A diff of the whole file for each edit |
+| Open files | Edits go through the editor, so open files update right away. The model reads what's in the editor, unsaved changes included |
+| Permission questions | **Allow**, **Always allow** (for this session) or **Reject**, in the editor's own dialog |
+| Modes | lcode's [permission modes](usage.md#permissions): Ask, Auto-edit, Plan and Yolo |
+| Plans | In Plan mode, the plan is approved in the editor, which then switches to Ask or Auto-edit |
+| The plan list | The model's todo list |
+| Context | How much of the context window is in use |
+| Commands | `/review` and [your own commands](commands.md) |
+| Past threads | lcode's saved sessions: an editor that reopens a thread gets its conversation back |
+
+[Checkpoints](usage.md#undo-and-checkpoints), [hooks and permission rules](hooks.md),
+[memory](memory.md), [subagents](agents.md) and [MCP servers](mcp.md) work as in the terminal.
+MCP servers the editor passes to its agents are added to lcode's own.
+
+## Settings
+
+lcode uses your [configuration](configuration.md): the model, the context window, the
+[model server](servers.md) and the default permission mode. To use different settings in an
+editor, set them in the agent's `env`, for example `"env": {"LCODE_MODEL": "qwen3.5-9b"}`.
+
+A repository's own commands, skills, agents and hooks are used only after you
+[approve them](commands.md#approving-a-repositorys-commands-skills-and-agents). An editor can't
+ask, so run `lcode` in the repository once and approve them there.
+
+To see lcode's own output while it runs in an editor, set `LCODE_ACP_LOG` to a file in `env`.

docs/usage.md

@@ -10,6 +10,7 @@ lcode doctor                 # check the installation (include this in bug repor
 lcode config [set|unset]     # show or change settings
 lcode bench [models]         # score models on small coding tasks on this machine
 lcode mcp [catalog|add|list] # connect MCP servers: Jira, GitHub, AWS, databases… (see MCP servers)
+lcode acp                    # run inside an editor over the Agent Client Protocol (see Editors)
 ```
 
 | Option | Meaning |

mkdocs.yml

@@ -97,6 +97,7 @@ nav:
       - Commands and skills: commands.md
       - Git workflow: git.md
       - Automation: automation.md
+      - Editors: editors.md
       - Code intelligence: lsp.md
       - Repository map and code search: search.md
       - Models & context windows: models.md

pyproject.toml

@@ -86,6 +86,7 @@ ignore = [
 "src/lcode/backends.py" = ["E501"]
 "src/lcode/gitflow.py" = ["E501"]
 "src/lcode/action.py" = ["E501"]
+"src/lcode/acp.py" = ["E501"]
 
 [tool.pytest.ini_options]
 testpaths = ["tests"]