elelem 0.8.0 → 0.9.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: a284087a47f32b75927bb0fc593aee12c376952f89af2ae61faa96b66810ae87
-  data.tar.gz: 138d5d8a0c110a1631c1a3b64024d2f82b5a80bfa1339244ef0d5bb932c2c626
+  metadata.gz: c88aa14330442939c8d62db38adcb229c860059888567b21fb309b760da18b7e
+  data.tar.gz: 7829340c692a3b025d95f47be9908c8d5ca0ab7988454b29e73af783d0e7d3ab
 SHA512:
-  metadata.gz: b3baefa781ccfef8e2a4882a1b8f65779b672f45918a9e0a9926741dbb80ecfb57c4c182f9074ae71e73b29010d3c349a893f3bde388d7e5340d04a158ae3cde
-  data.tar.gz: d4d6d0609f1ef0fa9bc5e9bb67aec6085944706db8a31025a63299f3ea0519f202a4a3f4a2ea4dad647bcb1390a98eb3ea7b1f0ab2f6b16c87ec18f6f1c355e3
+  metadata.gz: a3254b0655057dbaf8871ef6e6782bb962b588525a69a6d9f0746368f4c38d92c3ad41790104bb2cedfb25c16df2d77d07f236cfdde23805bee3be21d515cf99
+  data.tar.gz: f6d43cbe0659d37b6c6249bad2dc5c3dd04b6e4e2bb0a2a7201b4a479a9c1c52cc8e0516ffeb6b050098537e5c10b4d7887c80c8870ac01334599ab49100dc06

data/CHANGELOG.md

@@ -1,4 +1,73 @@
-## [Unreleased]
+## [0.9.0] - 2026-01-21
+
+### Added
+- **Plugin system** with support for custom tool definitions
+  - Load plugins from `lib/elelem/plugins/`, `~/.elelem/plugins/`, and `.elelem/plugins/`
+  - `Elelem::Plugins.register(name) { |toolbox| ... }` API
+  - Built-in plugins: `read`, `write`, `edit`, `execute`, `eval`, `verify`, `confirm`, `mcp`
+- **MCP (Model Context Protocol)** server support via `.mcp.json` configuration
+- **AGENTS.md** file support - searches up directory tree for project instructions
+- **`/init` command** to generate an AGENTS.md file for the current project
+- **`/shell` command** to drop into a shell session and capture the transcript to context
+- **`/reload` command** to hot-reload source code without restarting the process
+- **`task` tool** for delegating subtasks to focused sub-agents
+- **`edit` tool** for replacing first occurrence of text in a file
+- **`eval` tool** for executing Ruby code and dynamically registering new tools
+- **`verify` tool** for syntax checking and running project tests
+- **Pre/post tool hooks** (`toolbox.before`/`toolbox.after`) for extensibility
+- **Confirmation prompt** before executing shell commands (when TTY)
+- **Context compaction** for long conversations (summarizes old messages)
+- **Repo map** via ctags included in system prompt
+- **Markdown rendering** with [glow](https://github.com/charmbracelet/glow) for LLM responses
+- **CLI improvements**: optparse-based interface with `-p`/`-m` flags
+  - `elelem chat` - Interactive REPL (default)
+  - `elelem ask <prompt>` - One-shot query (reads stdin if piped)
+  - `elelem files` - Output files as XML
+- JSON Schema validation for tool call arguments (via `json_schemer`)
+- Tool aliases support (e.g., `bash`, `sh`, `exec` → `execute`)
+- **Dependencies documentation** in README with installation links
+
+### Changed
+- **Breaking**: Requires Ruby >= 4.0.0 (was 3.4.0)
+- **Breaking**: Removed `net-llm` dependency - LLM clients now inline in `lib/elelem/net/`
+  - `Elelem::Net::Claude` (Anthropic and Vertex AI)
+  - `Elelem::Net::OpenAI`
+  - `Elelem::Net::Ollama`
+- **Breaking**: Simplified LLM client `fetch` contract
+  - Yields `{content:, thinking:}` deltas
+  - Returns `tool_calls` array directly
+- **Breaking**: Tool schema uses OpenAI format (`{type: "function", function: {...}}`)
+- **Breaking**: Tool definitions use `description:` key (was `desc:`)
+- **Breaking**: Removed modes and permissions system entirely
+- **Breaking**: Removed slash commands (`/mode`, `/env`, `/provider`, `/model`)
+  - Remaining: `/clear`, `/context`, `/init`, `/reload`, `/shell`, `/exit`, `/help`
+- **Breaking**: Removed many dependencies
+  - Removed: `thor`, `cli-ui`, `erb`, `cgi`, `set`, `timeout`, `logger`, `net-llm`, `json-schema`
+  - Added: `json_schemer`, `optparse`, `tempfile`, `stringio`, `uri`
+- Consolidated multiple exe files into single `exe/elelem` entry point
+- Tools are now defined via plugins instead of hardcoded in Toolbox
+- System prompt includes hints for `rg`, `fd`, `sg` (ast-grep), `sed`, `patch`
+- System prompt regenerated on each fetch (includes dynamic repo map)
+- Default tool set: `read`, `write`, `edit`, `execute`, `eval`, `verify`, `task`
+- System prompt encourages using `eval` to create tools for repetitive tasks
+
+### Removed
+- `lib/elelem/application.rb` - CLI now in `exe/elelem`
+- `lib/elelem/conversation.rb` - simplified into Agent
+- `lib/elelem/git_context.rb` - inlined into Agent
+- `lib/elelem/system_prompt.erb` - now generated in Agent
+- `web_fetch`, `web_search`, `fetch`, `search_engine` tools
+- `patch` tool (use `edit` or `execute` with `sed`/`patch`)
+- `grep`, `list` tools (use `execute` with `rg`, `fd`)
+- Modes and permissions system
+- Events module
+- GitHub Actions CI workflow
+
+### Fixed
+- Handle missing args in Claude provider
+- Tool alias resolution (use canonical tool name, not alias)
+- Unknown tool error now suggests using `execute` and lists available tools
+- Duplicate write operations in edit flow
 
 ## [0.8.0] - 2026-01-14
 
@@ -48,7 +117,7 @@
 - Tab completion for `pass` entries without requiring `show` subcommand
 - Password store symlink support in tab completion
 
-## [0.5.0] - 2025-01-07
+## [0.5.0] - 2026-01-07
 
 ### Added
 - Multi-provider support: Ollama, Anthropic, OpenAI, and VertexAI
@@ -198,4 +267,3 @@
 ## [0.1.0] - 2025-08-08
 
 - Initial release
-

data/README.md

@@ -7,8 +7,7 @@ Fast, correct, autonomous – pick two.
 Elelem is a minimal coding agent written in Ruby. It is designed to help
 you write, edit, and manage code and plain-text files from the command line
 by delegating work to an LLM. The agent exposes a simple text-based UI and a
-set of built-in tools that give the LLM access to the local file system
-and Git.
+set of built-in tools that give the LLM access to the local file system.
 
 ## Design Principles
 
@@ -26,6 +25,26 @@ and Git.
 * Runs inside a Git repository.
 * Git is available and functional.
 
+## Dependencies
+
+Elelem relies on several external tools. Install the ones you need:
+
+| Tool | Purpose | Install |
+|------|---------|---------|
+| [Ollama](https://ollama.ai/) | Default LLM provider | https://ollama.ai/download |
+| [glow](https://github.com/charmbracelet/glow) | Markdown rendering | `brew install glow` / `go install github.com/charmbracelet/glow@latest` |
+| [ctags](https://ctags.io/) | Repo map generation | `brew install universal-ctags` / `apt install universal-ctags` |
+| [ripgrep](https://github.com/BurntSushi/ripgrep) | Text search (`rg`) | `brew install ripgrep` / `apt install ripgrep` |
+| [fd](https://github.com/sharkdp/fd) | File discovery | `brew install fd` / `apt install fd-find` |
+| [ast-grep](https://ast-grep.github.io/) | Structural search (`sg`) | `brew install ast-grep` / `cargo install ast-grep` |
+| [Git](https://git-scm.com/) | Version control | `brew install git` / `apt install git` |
+
+**Required:** Git, Ollama (or another LLM provider)
+
+**Recommended:** glow, ctags, ripgrep, fd
+
+**Optional:** ast-grep (for structural code search)
+
 ## Scope
 
 Only plain-text and source-code files are supported. No binary handling,
@@ -101,35 +120,10 @@ Each provider reads its configuration from environment variables:
 | openai      | `OPENAI_API_KEY`, `OPENAI_BASE_URL`               |
 | vertex-ai   | `GOOGLE_CLOUD_PROJECT`, `GOOGLE_CLOUD_REGION`     |
 
-## Mode System
-
-The agent exposes seven built‑in tools. You can switch which ones are
-available by changing the *mode*:
-
-| Mode    | Enabled Tools                            |
-|---------|------------------------------------------|
-| plan    | `grep`, `list`, `read`                   |
-| build   | `grep`, `list`, `read`, `patch`, `write` |
-| verify  | `grep`, `list`, `read`, `execute`        |
-| auto    | All tools                                |
-
-Use the following commands inside the REPL:
-
-```text
-/mode plan    # Read‑only
-/mode build   # Read + Write
-/mode verify  # Read + Execute
-/mode auto    # All tools
-/mode         # Show current mode
-```
-
-The system prompt is adjusted per mode so the LLM knows which actions
-are permissible.
-
 ## Features
 
 * **Interactive REPL** – clean, streaming chat.
-* **Toolbox** – file I/O, Git, shell execution.
+* **Toolbox** – file I/O and shell execution.
 * **Streaming Responses** – output appears in real time.
 * **Conversation History** – persists across turns; can be cleared.
 * **Context Dump** – `/context` shows the current conversation state.
@@ -137,24 +131,15 @@ are permissible.
 ## Toolbox Overview
 
 The `Toolbox` class is defined in `lib/elelem/toolbox.rb`. It supplies
-seven tools, each represented by a JSON schema that the LLM can call.
-
-| Tool      | Purpose                              | Parameters                           |
-| ----      | -------                              | ----------                           |
-| `exec`    | Run shell commands                   | `cmd`, `args`, `env`, `cwd`, `stdin` |
-| `eval`    | Dynamically create new tools         | `code`                               |
-| `grep`    | Search Git‑tracked files             | `query`                              |
-| `list`    | List tracked files                   | `path` (optional)                    |
-| `patch`   | Apply a unified diff via `git apply` | `diff`                               |
-| `read`    | Read file contents                   | `path`                               |
-| `write`   | Overwrite a file                     | `path`, `content`                    |
-
-## Tool Definition
-
-The core `Tool` wrapper is defined in `lib/elelem/tool.rb`. Each tool is
-created with a name, description, JSON schema for arguments, and a block
-that performs the operation. The LLM calls a tool by name and passes the
-arguments as a hash.
+three tools, each represented by a JSON schema that the LLM can call.
+
+| Tool      | Purpose            | Parameters         |
+| --------- | ------------------ | ------------------ |
+| `read`    | Read file contents | `path`             |
+| `write`   | Write file         | `path`, `content`  |
+| `execute` | Run shell command  | `command`          |
+
+Aliases: `bash`, `sh`, `exec` → `execute`; `open` → `read`
 
 ## Known Limitations
 

data/exe/elelem

@@ -2,9 +2,84 @@
 # frozen_string_literal: true
 
 require "elelem"
+require "optparse"
 
-Signal.trap("INT") do
-  exit(1)
+Signal.trap("INT") { exit 1 }
+
+class App
+  MODELS = {
+    "ollama" => "gpt-oss:latest",
+    "anthropic" => "claude-opus-4-5-20250514",
+    "vertex" => "claude-opus-4-5@20251101",
+    "openai" => "gpt-4o"
+  }.freeze
+
+  PROVIDERS = {
+    "ollama" => ->(model) { Elelem::Net::Ollama.new(model: model, host: ENV.fetch("OLLAMA_HOST", "localhost:11434")) },
+    "anthropic" => ->(model) { Elelem::Net::Claude.anthropic(model: model, api_key: ENV.fetch("ANTHROPIC_API_KEY")) },
+    "vertex" => ->(model) { Elelem::Net::Claude.vertex(model: model, project: ENV.fetch("GOOGLE_CLOUD_PROJECT"), region: ENV.fetch("GOOGLE_CLOUD_REGION", "us-east5")) },
+    "openai" => ->(model) { Elelem::Net::OpenAI.new(model: model, api_key: ENV.fetch("OPENAI_API_KEY")) }
+  }.freeze
+
+  def initialize(args)
+    @provider = "ollama"
+    @model = nil
+    @args = parse(args)
+  end
+
+  def run
+    command = @args.shift || "chat"
+    send(command.tr("-", "_"))
+  rescue NoMethodError
+    abort "Unknown command: #{command}"
+  end
+
+  private
+
+  def parse(args)
+    @parser = OptionParser.new do |o|
+      o.banner = "Usage: elelem [command] [options] [args]"
+      o.separator "\nCommands:"
+      o.separator "  chat              Interactive REPL (default)"
+      o.separator "  ask <prompt>      One-shot query (reads stdin if piped)"
+      o.separator "  files             Output files as XML (no options)"
+      o.separator "  help              Show this help"
+      o.separator "\nOptions:"
+      o.on("-p", "--provider NAME", "ollama, anthropic, vertex, openai") { |p| @provider = p }
+      o.on("-m", "--model NAME", "Override default model") { |m| @model = m }
+      o.on("-h", "--help") { puts o; exit }
+    end
+    @parser.parse!(args)
+  end
+
+  def help
+    puts @parser
+  end
+
+  def client
+    model = @model || MODELS.fetch(@provider)
+    PROVIDERS.fetch(@provider).call(model)
+  end
+
+  def chat = Elelem.start(client)
+
+  def ask
+    abort "Usage: elelem ask <prompt>" if @args.empty?
+    prompt = @args.join(" ")
+    prompt = "#{prompt}\n\n```\n#{$stdin.read}\n```" if $stdin.stat.pipe?
+    Elelem::Terminal.new.markdown Elelem.ask(client, prompt)
+  end
+
+  def files
+    files = $stdin.stat.pipe? ? $stdin.readlines : `git ls-files`.lines
+    puts "<documents>"
+    files.each_with_index do |line, i|
+      path = line.strip
+      next if path.empty? || !File.file?(path)
+      puts %Q{<document index="#{i + 1}"><source>#{path}</source><content><![CDATA[#{File.read(path)}]]></content></document>}
+    end
+    puts "</documents>"
+  end
 end
 
-Elelem::Application.start
+App.new(ARGV).run