feat: add agentic workflows to the plugin

README.md

@@ -16,7 +16,7 @@ Currently supports: Anthropic, Ollama and OpenAI adapters
 </p>
 
 > [!IMPORTANT]
-> This plugin is provided as-is and is primarily developed for my own workflows. As such, I offer no guarantees of regular updates or support and I expect the plugin's API to change regularly. Bug fixes and feature enhancements will be implemented at my discretion, and only if they align with my personal use-case. Feel free to fork the project and customize it to your needs, but please understand my involvement in further development will be intermittent. To be notified of breaking changes in the plugin, please subscribe to [this issue](https://github.com/olimorris/codecompanion.nvim/issues/9).
+> This plugin is provided as-is and is primarily developed for my own workflows. As such, I offer no guarantees of regular updates or support and I expect the plugin's API to change regularly. Bug fixes and feature enhancements will be implemented at my discretion, and only if they align with my personal use-cases. Feel free to fork the project and customize it to your needs, but please understand my involvement in further development will be intermittent. To be notified of breaking changes in the plugin, please subscribe to [this issue](https://github.com/olimorris/codecompanion.nvim/issues/9).
 
 <p align="center">
 <img src="https://github.com/olimorris/codecompanion.nvim/assets/9512444/5e5a5e54-c1d9-4fe2-8ae0-1cfbfdd6cea5" alt="Header" />
@@ -28,6 +28,7 @@ Currently supports: Anthropic, Ollama and OpenAI adapters
 
 - :speech_balloon: A Copilot Chat experience from within Neovim
 - :electric_plug: Adapter support for many generative AI services
+- :robot: Agentic workflows to improve LLM output
 - :rocket: Inline code creation and modification
 - :sparkles: Built in actions for specific language prompts, LSP error fixes and code advice
 - :building_construction: Create your own custom actions for Neovim
@@ -245,13 +246,16 @@ The plugin has a number of commands:
 - `:CodeCompanionToggle` - Toggle a chat buffer
 - `:CodeCompanionActions` - To open up the action palette window
 
-For an optimum workflow, the plugin author recommendeds the following keymaps:
+For an optimum workflow, the plugin author recommendeds the following:
 
 ```lua
 vim.api.nvim_set_keymap("n", "<C-a>", "<cmd>CodeCompanionActions<cr>", { noremap = true, silent = true })
 vim.api.nvim_set_keymap("v", "<C-a>", "<cmd>CodeCompanionActions<cr>", { noremap = true, silent = true })
 vim.api.nvim_set_keymap("n", "<LocalLeader>a", "<cmd>CodeCompanionToggle<cr>", { noremap = true, silent = true })
 vim.api.nvim_set_keymap("v", "<LocalLeader>a", "<cmd>CodeCompanionToggle<cr>", { noremap = true, silent = true })
+
+-- Expand `cc` into CodeCompanion in the command line
+vim.cmd([[cab cc CodeCompanion]])
 ```
 
 > [!NOTE]
@@ -329,17 +333,34 @@ The plugin comes with a number of [in-built actions](https://github.com/olimorri
 
 #### Chat and Chat as
 
-Both of these actions utilise the `chat` strategy. The `Chat` action opens up a fresh chat buffer. The `Chat as` action allows for persona based context to be set in the chat buffer allowing for better and more detailed responses from the generative AI service.
-
 > [!TIP]
 > Both of these actions allow for visually selected code to be sent to the chat buffer as code blocks.
 
+Both of these actions utilise the `chat` strategy. The `Chat` action opens up a fresh chat buffer. The `Chat as` action allows for persona based context to be set in the chat buffer allowing for better and more detailed responses from the generative AI service.
+
 #### Open chats
 
 This action enables users to easily navigate between their open chat buffers. A chat buffer can be deleted (and removed from memory) by pressing `<C-c>`.
 
+#### Agentic Workflows
+
+> [!WARNING]
+> Agentic workflows may result in the significant consumption of tokens if you're using an external generative AI service.
+
+As outlined in Andrew Ng's [tweet](https://twitter.com/AndrewYNg/status/1773393357022298617), agentic workflows have the ability to dramatically improve the output of an LLM and can be as simple as prompting an LLM multiple times. The plugin supports this via the use of workflows. At various stages of the workflow, the plugin will automatically prompt the LLM for feedback and self-reflection without any input from the user.
+
+Currently, the plugin only supports _"reflection"_ (multiple prompts within the same application) and comes with the following workflows:
+
+- Adding a new feature
+- Refactoring code
+
+Of course you can add new workflows by following the [RECIPES](RECIPES.md) guide.
+
 #### Inline code
 
+> [!NOTE]
+> The options available to the user in the Action Palette will depend on the Vim mode.
+
 These actions utilize the `inline` strategy. They can be useful for writing inline code in a buffer or even refactoring a visual selection; all based on a user's prompt. The actions are designed to write code for the buffer filetype that it is initated in, or, if run from a terminal prompt, to write commands.
 
 The strategy comes with a number of helpers which the user can type in the prompt, similar to [GitHub Copilot Chat](https://github.blog/changelog/2024-01-30-code-faster-and-better-with-github-copilots-new-features-in-visual-studio/):
@@ -348,15 +369,19 @@ The strategy comes with a number of helpers which the user can type in the promp
 - `/optimize` to analyze and improve the running time of the selected code
 - `/tests` to create unit tests for the selected code
 
-> [!NOTE]
-> The options available to the user in the Action Palette will depend on the Vim mode.
 
 #### Code advisor
 
+> [!NOTE]
+> This option is only available in visual mode
+
 As the name suggests, this action provides advice on a visual selection of code and utilises the `chat` strategy. The response from the API is streamed into a chat buffer which follows the `display.chat` settings in your configuration.
 
 #### LSP assistant
 
+> [!NOTE]
+> This option is only available in visual mode
+
 Taken from the fantastic [Wtf.nvim](https://github.com/piersolenski/wtf.nvim) plugin, this action provides advice on how to correct any LSP diagnostics which are present on the visually selected lines. Again, the `send_code = false` value can be set in your config to prevent the code itself being sent to the generative AI service.
 
 ## :rainbow: Helpers

doc/codecompanion.txt

@@ -1,4 +1,4 @@
-*codecompanion.txt*         For NVIM v0.9.2         Last change: 2024 March 24
+*codecompanion.txt*         For NVIM v0.9.2         Last change: 2024 April 01
 
 ==============================================================================
 Table of Contents                            *codecompanion-table-of-contents*
@@ -14,6 +14,7 @@ FEATURES                                              *codecompanion-features*
 
 - A Copilot Chat experience from within Neovim
 - Adapter support for many generative AI services
+- Agentic workflows to improve LLM output
 - Inline code creation and modification
 - Built in actions for specific language prompts, LSP error fixes and code advice
 - Create your own custom actions for Neovim
@@ -234,13 +235,16 @@ The plugin has a number of commands:
 - `:CodeCompanionToggle` - Toggle a chat buffer
 - `:CodeCompanionActions` - To open up the action palette window
 
-For an optimum workflow, the plugin author recommendeds the following keymaps:
+For an optimum workflow, the plugin author recommendeds the following:
 
 >lua
     vim.api.nvim_set_keymap("n", "<C-a>", "<cmd>CodeCompanionActions<cr>", { noremap = true, silent = true })
     vim.api.nvim_set_keymap("v", "<C-a>", "<cmd>CodeCompanionActions<cr>", { noremap = true, silent = true })
     vim.api.nvim_set_keymap("n", "<LocalLeader>a", "<cmd>CodeCompanionToggle<cr>", { noremap = true, silent = true })
     vim.api.nvim_set_keymap("v", "<LocalLeader>a", "<cmd>CodeCompanionToggle<cr>", { noremap = true, silent = true })
+
+    -- Expand `cc` into CodeCompanion in the command line
+    vim.cmd([[cab cc CodeCompanion]])
 <
 
 
@@ -353,23 +357,49 @@ Neovim buffer.
 
 CHAT AND CHAT AS
 
+
+  [!TIP] Both of these actions allow for visually selected code to be sent to the
+  chat buffer as code blocks.
 Both of these actions utilise the `chat` strategy. The `Chat` action opens up a
 fresh chat buffer. The `Chat as` action allows for persona based context to be
 set in the chat buffer allowing for better and more detailed responses from the
 generative AI service.
 
 
-  [!TIP] Both of these actions allow for visually selected code to be sent to the
-  chat buffer as code blocks.
-
 OPEN CHATS
 
 This action enables users to easily navigate between their open chat buffers. A
 chat buffer can be deleted (and removed from memory) by pressing `<C-c>`.
 
 
+AGENTIC WORKFLOWS
+
+
+  [!WARNING] Agentic workflows may result in the significant consumption of
+  tokens if you’re using an external generative AI service.
+As outlined in Andrew Ng’s tweet
+<https://twitter.com/AndrewYNg/status/1773393357022298617>, agentic workflows
+have the ability to dramatically improve the output of an LLM and can be as
+simple as prompting an LLM multiple times. The plugin supports this via the use
+of workflows. At various stages of the workflow, the plugin will automatically
+prompt the LLM for feedback and self-reflection without any input from the
+user.
+
+Currently, the plugin only supports _“reflection”_ (multiple prompts within
+the same application) and comes with the following workflows:
+
+- Adding a new feature
+- Refactoring code
+
+Of course you can add new workflows by following the RECIPES <RECIPES.md>
+guide.
+
+
 INLINE CODE
 
+
+  [!NOTE] The options available to the user in the Action Palette will depend on
+  the Vim mode.
 These actions utilize the `inline` strategy. They can be useful for writing
 inline code in a buffer or even refactoring a visual selection; all based on a
 user’s prompt. The actions are designed to write code for the buffer filetype
@@ -384,18 +414,19 @@ prompt, similar to GitHub Copilot Chat
 - `/tests` to create unit tests for the selected code
 
 
-  [!NOTE] The options available to the user in the Action Palette will depend on
-  the Vim mode.
-
 CODE ADVISOR
 
+
+  [!NOTE] This option is only available in visual mode
 As the name suggests, this action provides advice on a visual selection of code
 and utilises the `chat` strategy. The response from the API is streamed into a
 chat buffer which follows the `display.chat` settings in your configuration.
 
 
 LSP ASSISTANT
 
+
+  [!NOTE] This option is only available in visual mode
 Taken from the fantastic Wtf.nvim <https://github.com/piersolenski/wtf.nvim>
 plugin, this action provides advice on how to correct any LSP diagnostics which
 are present on the visually selected lines. Again, the `send_code = false`

lua/codecompanion/actions.lua

@@ -289,6 +289,124 @@ M.static.actions = {
       },
     },
   },
+  {
+    name = "Agentic Workflows...",
+    strategy = "chat",
+    description = "Workflows to improve the performance of your LLM",
+    picker = {
+      prompt = "Select a workflow",
+      items = {
+        {
+          name = "Code a feature - Outline, draft, consider and then revise",
+          callback = function(context)
+            local agent = require("codecompanion.agent")
+            return agent
+              .new({
+                context = context,
+                strategy = "chat",
+              })
+              :workflow({
+                {
+                  role = "system",
+                  content = "You are an expert coder and helpful assistant who can help outline, draft, consider and revise code for the "
+                    .. context.filetype
+                    .. " language.",
+                  start = true,
+                },
+                {
+                  condition = function()
+                    return context.is_visual
+                  end,
+                  contains_code = true,
+                  role = "user",
+                  content = "Here is some relevant context: " .. send_code(context),
+                  start = true,
+                },
+                {
+                  role = "user",
+                  content = "I want you to help me code a feature. Before we write any code let's outline how we'll architect and implement the feature with the context you already have. The feature I'd like to add is ",
+                  start = true,
+                },
+                {
+                  role = "user",
+                  content = "Thanks. Now let's draft the code for the feature.",
+                  auto_submit = true,
+                },
+                {
+                  role = "user",
+                  content = "Great. Now let's consider the code. I'd like you to check it carefully for correctness, style, and efficiency, and give constructive criticism for how to improve it.",
+                  auto_submit = true,
+                },
+                {
+                  role = "user",
+                  content = "Thanks. Now let's revise the code based on the feedback.",
+                  auto_submit = true,
+                },
+                {
+                  role = "user",
+                  content = "For clarity, can you show the final code without any explanations?",
+                  auto_submit = true,
+                },
+              })
+          end,
+        },
+        {
+          name = "Refactor some code - Outline, draft, consider and then revise",
+          callback = function(context)
+            local agent = require("codecompanion.agent")
+            return agent
+              .new({
+                context = context,
+                strategy = "chat",
+              })
+              :workflow({
+                {
+                  role = "system",
+                  content = "You are an expert coder and helpful assistant who can help outline, draft, consider and revise code for the "
+                    .. context.filetype
+                    .. " language.",
+                  start = true,
+                },
+                {
+                  condition = function()
+                    return context.is_visual
+                  end,
+                  contains_code = true,
+                  role = "user",
+                  content = "Here is some relevant context: " .. send_code(context),
+                  start = true,
+                },
+                {
+                  role = "user",
+                  content = "I want you to help me with a refactor. Before we write any code let's outline how we'll architect and implement the code with the context you already have. What I'm looking to achieve is ",
+                  start = true,
+                },
+                {
+                  role = "user",
+                  content = "Thanks. Now let's draft the code for the refactor.",
+                  auto_submit = true,
+                },
+                {
+                  role = "user",
+                  content = "Great. Now let's consider the code. I'd like you to check it carefully for correctness, style, and efficiency, and give constructive criticism for how to improve it.",
+                  auto_submit = true,
+                },
+                {
+                  role = "user",
+                  content = "Thanks. Now let's revise the code based on the feedback.",
+                  auto_submit = true,
+                },
+                {
+                  role = "user",
+                  content = "For clarity, can you show the final code without any explanations?",
+                  auto_submit = true,
+                },
+              })
+          end,
+        },
+      },
+    },
+  },
   {
     name = "Inline code ...",
     strategy = "inline",

lua/codecompanion/agent.lua

@@ -0,0 +1,52 @@
+local config = require("codecompanion.config")
+
+---@class CodeCompanion.Agent
+local Agent = {}
+
+---@class CodeCompanion.AgentArgs
+---@field context table
+---@field strategy string
+
+---@param args table
+---@return CodeCompanion.Agent
+function Agent.new(args)
+  return setmetatable(args, { __index = Agent })
+end
+
+---@param prompts table
+function Agent:workflow(prompts)
+  local starting_prompts = {}
+  local workflow_prompts = {}
+
+  for _, prompt in ipairs(prompts) do
+    if prompt.start then
+      if
+        (type(prompt.condition) == "function" and not prompt.condition())
+        or (prompt.contains_code and not config.options.send_code)
+      then
+        goto continue
+      end
+
+      table.insert(starting_prompts, {
+        role = prompt.role,
+        content = prompt.content,
+      })
+    else
+      table.insert(workflow_prompts, {
+        role = prompt.role,
+        content = prompt.content,
+        auto_submit = prompt.auto_submit,
+      })
+    end
+    ::continue::
+  end
+
+  return require("codecompanion.strategies.chat").new({
+    type = "chat",
+    messages = starting_prompts,
+    workflow = workflow_prompts,
+    show_buffer = true,
+  })
+end
+
+return Agent

lua/codecompanion/init.lua

@@ -97,7 +97,9 @@ M.actions = function(args)
       format = function(item)
         local formatted_item = {}
         for _, column in ipairs(opts.columns) do
-          table.insert(formatted_item, item[column] or "")
+          if item[column] ~= nil then
+            table.insert(formatted_item, item[column] or "")
+          end
         end
         return formatted_item
       end,
@@ -119,7 +121,7 @@ M.actions = function(args)
       }
       picker(actions.validate(item.picker.items(), context), picker_opts, selection)
     elseif item and type(item.callback) == "function" then
-      return item.callback(selection)
+      return item.callback(context)
     else
       local Strategy = require("codecompanion.strategy")
       return Strategy.new({