roast-ai 0.3.0 → 0.4.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 28055339ac18d63a1abd7151943558a114df2e48930c63aa63c65536d193061b
-  data.tar.gz: c5be17b70caa32f73dca316ab4816d0b92c0c7121747b3e5492f58124e815f87
+  metadata.gz: 0c0f1a837c5dafcaecb8443e93a4076c11566f1c3bea1d6835121628fbdf79a7
+  data.tar.gz: ab3679b374486831cbf90ab3a259178d48d4e53d5d5dbdc2a2d43537399e4ca9
 SHA512:
-  metadata.gz: d5b300fd001d8cc55ecdb5751bc891840732f63be8e951e2c4333a178fde6c2197a2b0063f3add45b48e85251e9b2ee28119d237322bb8f89af86c2208345c1f
-  data.tar.gz: c579010aedef591cd1a72fb6cd34c15db38fc225550aa17a841397e5f8849c09c5f87cf81da3f04f2300c5dcac704ab2d8a0f78fe8fd443fabfe62e67780a015
+  metadata.gz: 3061d9f04b7f886e3a6d588eb729e10062f6310d5973c7ed5181975f27a385c20d8ff8880ecb7e61abd54dfbd87795211313e1c9a6d2a630269d7c4559ae7cae
+  data.tar.gz: 7962e4991590700003f3f8a827435214ef9aa90fa4324d34d3b421f11d5f647e43663e17189a0b43f81d6fa124bbf68a269f9674a493994832b9ab5c03f192a6

data/.gitignore

@@ -40,3 +40,4 @@ bin/ruby-rewrite
 bin/thor
 
 gemfiles/*.lock
+bin/claude-swarm

data/CHANGELOG.md

@@ -5,6 +5,38 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [0.4.0] - 2025-06-12
+
+### Added
+- **Input step type** for collecting user input during workflow execution (#154)
+  - Interactive prompts pause workflow execution to collect user input
+  - Supports multiple input types: `text` (default), `confirm`, `select`, and `password`
+  - `confirm` type provides yes/no prompts with boolean results
+  - `select` type allows choosing from a list of options
+  - `password` type masks input for sensitive data using io/console
+  - Input values are stored in workflow output and accessible via dot notation (e.g., `{{output.step_name}}`)
+  - Integrates with CLI::UI for consistent formatting and user experience
+- **Agent step type** for direct pass-through to coding agents (#151)
+  - Steps prefixed with `^` send prompts directly to the CodingAgent tool
+  - Supports both file-based and inline agent prompts
+  - Bypasses LLM interpretation for precise agent instructions
+
+### Fixed
+- DotAccessHash array wrapping and template response handling
+- CLI::UI formatting and color handling for better terminal output
+
+## [0.3.1] - 2025-06-05
+
+### Added
+- Default `print_response: true` for the last step in a workflow (#100)
+  - The last step now automatically prints its response unless explicitly configured otherwise
+  - Helps newcomers who would otherwise see no output from their workflows
+  - Works with all step types: string steps, hash steps with variable assignment, and conditional steps
+  - Parallel steps and iteration steps are intelligently handled (no automatic output since there's no single "last" step)
+
+### Fixed
+- PromptStep now properly passes `print_response`, `json`, and `params` parameters to chat_completion
+
 ## [0.3.0] - 2025-06-04
 
 ### Changed

data/CLAUDE.md

@@ -4,6 +4,7 @@ This file provides guidance to Claude Code (claude.ai/code) when working with co
 
 ## About the codebase
 - This is a Ruby gem called Roast. Its purpose is to run AI workflows defined in a YAML file.
+- Note that this project now uses Zeitwerk, which means you don't have to manually require project files anymore
 
 ## Commands
 
@@ -51,11 +52,14 @@ This file provides guidance to Claude Code (claude.ai/code) when working with co
   - ConditionalExecutor handles conditionals (if, unless)
   - Don't combine different responsibilities in one class
 - **Do not implement prompts "inline" using a prompt: attribute nested under step names, that violates the primary design architecture of Roast**
+- When faced with the choice between working around an architectural issue or code smell versus actually diving into fixing the design issue or code smell, choose the latter more principled approach
+- When fixing code smells, you don't have to worry about internal backwards compatibility 
 
 ## Guidance and Expectations
 
 - Do not decide unilaterally to leave code for the sake of "backwards compatibility"... always run those decisions by me first.
 - Don't ever commit and push changes unless directly told to do so
+- You can't test input steps yourself, ask me to do it
 
 ## Git Workflow Practices
 
@@ -130,6 +134,48 @@ gh pr view {pr_number}
 gh pr diff {pr_number}
 ```
 
+### Issue Labeling and Project Management
+
+When creating GitHub issues, always check available labels, projects, and milestones first:
+
+```bash
+# List all available labels
+gh api repos/Shopify/roast/labels | jq '.[].name'
+
+# List all milestones
+gh api repos/Shopify/roast/milestones | jq '.[] | {title: .title, number: .number, state: .state}'
+
+# List projects linked to the roast repository
+gh api graphql -f query='
+{
+  repository(owner: "Shopify", name: "roast") {
+    projectsV2(first: 10) {
+      nodes {
+        title
+        number
+        url
+      }
+    }
+  }
+}' --jq '.data.repository.projectsV2.nodes[] | {title: .title, number: .number, url: .url}'
+```
+
+**Issue Creation Workflow**:
+1. First check what labels exist and apply appropriate ones when creating the issue
+2. After creating the issue, ask the user if they want it added to an existing milestone
+3. Ask the user if they want it added to a particular project board
+
+```bash
+# Create issue with labels
+gh api repos/Shopify/roast/issues -X POST -F title="Issue Title" -F body="Issue description" -F 'labels=["bug", "enhancement"]'
+
+# Add issue to a milestone (after creation)
+gh api repos/Shopify/roast/issues/{issue_number} -X PATCH -F milestone={milestone_number}
+
+# Add issue to a GitHub Project
+gh project item-add {project_number} --url https://github.com/Shopify/roast/issues/{issue_number}
+```
+
 #### Formatting Tips for GitHub API
 1. Use literal newlines in the body text instead of `\n` escape sequences
 2. When formatting is stripped (like backticks), use alternatives:
@@ -167,4 +213,9 @@ gh pr diff {pr_number}
     - Avoid premature optimization outside of hot paths
     - Consider the tradeoff between readability and performance
     - Suggest optimizations that improve both clarity and performance
-```
+
+## CLI::UI Formatting Tips
+- To apply color to terminal output using CLI::UI, use the following syntax:
+  ```ruby
+  puts ::CLI::UI.fmt("{{red:This field is required. Please provide a value.}}")
+  ```

data/Gemfile

@@ -8,7 +8,7 @@ git_source(:github) { |repo_name| "https://github.com/#{repo_name}" }
 gemspec
 
 gem "cgi"
-gem "cli-ui"
+gem "cli-ui", "2.3.0"
 gem "dotenv"
 gem "guard"
 gem "guard-minitest"
@@ -18,3 +18,5 @@ gem "rubocop-shopify", require: false
 gem "vcr", require: false
 gem "webmock", require: false
 gem "minitest-rg"
+
+gem "claude_swarm"

data/Gemfile.lock

@@ -1,15 +1,16 @@
 PATH
   remote: .
   specs:
-    roast-ai (0.3.0)
+    roast-ai (0.4.0)
       activesupport (>= 7.0)
-      cli-ui
+      cli-ui (= 2.3.0)
       diff-lcs (~> 1.5)
       faraday-retry
       json-schema
       open_router (~> 0.3)
       raix (~> 1.0)
       thor (~> 1.3)
+      zeitwerk (~> 2.6)
 
 GEM
   remote: https://rubygems.org/
@@ -33,17 +34,49 @@ GEM
     base64 (0.3.0)
     benchmark (0.4.1)
     bigdecimal (3.2.2)
-    cgi (0.4.2)
-    cli-ui (2.3.1)
+    cgi (0.5.0)
+    claude_swarm (0.1.15)
+      fast-mcp-annotations
+      thor (~> 1.3)
+    cli-ui (2.3.0)
     coderay (1.1.3)
     concurrent-ruby (1.3.5)
     connection_pool (2.5.3)
     crack (1.0.0)
       bigdecimal
       rexml
-    diff-lcs (1.6.1)
+    diff-lcs (1.6.2)
     dotenv (3.1.8)
     drb (2.2.3)
+    dry-configurable (1.3.0)
+      dry-core (~> 1.1)
+      zeitwerk (~> 2.6)
+    dry-core (1.1.0)
+      concurrent-ruby (~> 1.0)
+      logger
+      zeitwerk (~> 2.6)
+    dry-inflector (1.2.0)
+    dry-initializer (3.2.0)
+    dry-logic (1.6.0)
+      bigdecimal
+      concurrent-ruby (~> 1.0)
+      dry-core (~> 1.1)
+      zeitwerk (~> 2.6)
+    dry-schema (1.14.1)
+      concurrent-ruby (~> 1.0)
+      dry-configurable (~> 1.0, >= 1.0.1)
+      dry-core (~> 1.1)
+      dry-initializer (~> 3.2)
+      dry-logic (~> 1.5)
+      dry-types (~> 1.8)
+      zeitwerk (~> 2.6)
+    dry-types (1.8.3)
+      bigdecimal (~> 3.0)
+      concurrent-ruby (~> 1.0)
+      dry-core (~> 1.0)
+      dry-inflector (~> 1.0)
+      dry-logic (~> 1.4)
+      zeitwerk (~> 2.6)
     event_stream_parser (1.0.0)
     faraday (2.13.1)
       faraday-net_http (>= 2.0, < 3.5)
@@ -55,6 +88,13 @@ GEM
       net-http (>= 0.5.0)
     faraday-retry (2.3.1)
       faraday (~> 2.0)
+    fast-mcp-annotations (1.5.2)
+      addressable (~> 2.8)
+      base64
+      dry-schema (~> 1.14)
+      json (~> 2.0)
+      mime-types (~> 3.4)
+      rack (~> 3.1)
     ffi (1.17.2-arm64-darwin)
     ffi (1.17.2-x86_64-linux-gnu)
     formatador (1.1.0)
@@ -73,14 +113,14 @@ GEM
     guard-minitest (2.4.6)
       guard-compat (~> 1.2)
       minitest (>= 3.0)
-    hashdiff (1.1.2)
+    hashdiff (1.2.0)
     i18n (1.14.7)
       concurrent-ruby (~> 1.0)
     json (2.12.2)
     json-schema (5.1.1)
       addressable (~> 2.8)
       bigdecimal (~> 3.1)
-    language_server-protocol (3.17.0.4)
+    language_server-protocol (3.17.0.5)
     lint_roller (1.1.0)
     listen (3.9.0)
       rb-fsevent (~> 0.10, >= 0.10.3)
@@ -88,6 +128,10 @@ GEM
     logger (1.7.0)
     lumberjack (1.2.10)
     method_source (1.1.0)
+    mime-types (3.7.0)
+      logger
+      mime-types-data (~> 3.2025, >= 3.2025.0507)
+    mime-types-data (3.2025.0603)
     minitest (5.25.5)
     minitest-rg (5.3.0)
       minitest (~> 5.0)
@@ -114,22 +158,23 @@ GEM
     pry (0.15.2)
       coderay (~> 1.1)
       method_source (~> 1.0)
-    public_suffix (6.0.1)
+    public_suffix (6.0.2)
     racc (1.8.1)
+    rack (3.1.16)
     rainbow (3.1.1)
-    raix (1.0.0)
+    raix (1.0.1)
       activesupport (>= 6.0)
       faraday-retry (~> 2.0)
       open_router (~> 0.2)
       ostruct
       ruby-openai (~> 7)
-    rake (13.2.1)
+    rake (13.3.0)
     rb-fsevent (0.11.2)
     rb-inotify (0.11.1)
       ffi (~> 1.0)
     regexp_parser (2.10.0)
     rexml (3.4.1)
-    rubocop (1.75.3)
+    rubocop (1.76.0)
       json (~> 2.3)
       language_server-protocol (~> 3.17.0.2)
       lint_roller (~> 1.1.0)
@@ -137,13 +182,13 @@ GEM
       parser (>= 3.3.0.2)
       rainbow (>= 2.2.2, < 4.0)
       regexp_parser (>= 2.9.3, < 3.0)
-      rubocop-ast (>= 1.44.0, < 2.0)
+      rubocop-ast (>= 1.45.0, < 2.0)
       ruby-progressbar (~> 1.7)
       unicode-display_width (>= 2.4.0, < 4.0)
-    rubocop-ast (1.44.1)
+    rubocop-ast (1.45.0)
       parser (>= 3.3.7.2)
       prism (~> 1.4)
-    rubocop-shopify (2.17.0)
+    rubocop-shopify (2.17.1)
       rubocop (~> 1.62)
     ruby-openai (7.4.0)
       event_stream_parser (>= 0.3.0, < 2.0.0)
@@ -166,14 +211,16 @@ GEM
       addressable (>= 2.8.0)
       crack (>= 0.3.2)
       hashdiff (>= 0.4.0, < 2.0.0)
+    zeitwerk (2.7.3)
 
 PLATFORMS
-  arm64-darwin-23
+  arm64-darwin
   x86_64-linux
 
 DEPENDENCIES
   cgi
-  cli-ui
+  claude_swarm
+  cli-ui (= 2.3.0)
   dotenv
   guard
   guard-minitest

data/README.md

@@ -82,13 +82,13 @@ steps:
 
 ## Try it
 
-If you don't have one already, get an OpenAI key from [here](https://platform.openai.com/settings/organization/api-keys). You will need an account with a credit card, make sure that a basic completion works.
+If you don’t have one already, get an OpenAI key from [here](https://platform.openai.com/settings/organization/api-keys). You will need an account with a credit card and credits applied to the associated project. Make sure that a basic completion works:
 
 ```bash
 export OPENAI_API_KEY=sk-proj-....
 
 curl -H "Content-Type: application/json" \
-    -H "Authorization: Bearer $API_TOKEN" \
+    -H "Authorization: Bearer $OPENAI_API_KEY" \
     -d '{"model":"gpt-4.1-mini","messages":[{"role":"user","content":"What is 1+1?"}]}' \
     https://api.openai.com/v1/chat/completions
 ```
@@ -270,6 +270,70 @@ Roast supports several types of steps:
    ```
    This creates a simple prompt-response interaction without tool calls or looping. It's detected by the presence of spaces in the step name and is useful for summarization or simple questions at the end of a workflow.
 
+8. **Agent step**: Direct pass-through to coding agents (e.g., Claude Code)
+   ```yaml
+   steps:
+     - ^fix_linting_errors                                    # File-based agent prompt
+     - ^Review the code and identify any performance issues   # Inline agent prompt
+     - regular_analysis                                       # Normal step through LLM
+   ```
+   Agent steps are prefixed with `^` and send the prompt content directly to the CodingAgent tool without LLM translation. This is useful when you want to give precise instructions to a coding agent without the intermediate interpretation layer. Agent steps support both file-based prompts (`fix_linting_errors/prompt.md`) and inline prompts (text with spaces).
+
+9. **Input step**: Interactive prompts for user input during workflow execution
+   ```yaml
+   steps:
+     - analyze_code
+     - get_user_feedback:
+         prompt: "Should we proceed with the refactoring? (yes/no)"
+         type: confirm
+     - review_changes:
+         prompt: "Enter your review comments"
+         type: text
+     - select_strategy:
+         prompt: "Choose optimization strategy"
+         type: select
+         options:
+           - "Performance optimization"
+           - "Memory optimization"
+           - "Code clarity"
+     - api_configuration:
+         prompt: "Enter API key"
+         type: password
+   ```
+   
+   Input steps pause workflow execution to collect user input. They support several types:
+   - `text`: Free-form text input (default if type not specified)
+   - `confirm`: Yes/No confirmation prompts
+   - `select`: Choice from a list of options
+   - `password`: Masked input for sensitive data
+   
+   The user's input is stored in the workflow output using the step name as the key and can be accessed in subsequent steps via interpolation (e.g., `{{output.get_user_feedback}}`).
+
+#### Step Configuration
+
+Steps can be configured with various options to control their behavior:
+
+```yaml
+steps:
+  - analyze_code           # Simple step reference
+  - generate_report:       # Step with configuration
+      model: gpt-4o        # Override the global model for this step
+      print_response: true # Explicitly control output printing
+      json: true           # Request JSON-formatted response
+      params:              # Additional parameters for the API call
+        temperature: 0.8
+```
+
+**Configuration options:**
+- `model`: Override the workflow's default model for this specific step
+- `print_response`: Control whether the step's response is included in the final output (default: `false`, except for the last step which defaults to `true` as of v0.3.1)
+- `json`: Request a JSON-formatted response from the model
+- `params`: Additional parameters passed to the model API (temperature, max_tokens, etc.)
+- `path`: Custom directory path for the step's prompt files
+- `coerce_to`: Type coercion for the step result (`:boolean`, `:llm_boolean`, `:iterable`)
+
+**Automatic Last Step Output**: As of version 0.3.1, the last step in a workflow automatically has `print_response: true` unless explicitly configured otherwise. This ensures that newcomers to Roast see output from their workflows by default.
+
 #### Shared Configuration
 
 Roast supports sharing common configuration and steps across multiple workflows using a `shared.yml` file.
@@ -633,6 +697,52 @@ tools:
 
 Custom descriptions help the LLM understand when and how to use each command, making your workflows more effective.
 
+### Step-Level Tool Filtering
+
+You can restrict which tools are available to specific steps using the `available_tools` configuration:
+
+```yaml
+# Define all tools globally
+tools:
+  - Roast::Tools::Grep
+  - Roast::Tools::ReadFile
+  - Roast::Tools::WriteFile
+  - Roast::Tools::Cmd:
+      allowed_commands:
+        - pwd
+        - ls
+        - echo
+
+# Configure steps with specific tool access
+explore_directory:
+  available_tools:
+    - pwd
+    - ls
+
+analyze_files:
+  available_tools:
+    - grep
+    - read_file
+
+write_summary:
+  available_tools:
+    - write_file
+    - echo
+```
+
+This feature provides:
+- **Security**: Each step only has access to the tools it needs
+- **Performance**: Reduces the tool list sent to the LLM
+- **Clarity**: Makes tool usage explicit for each step
+
+Key points:
+- Use snake_case tool names (e.g., `read_file` for `Roast::Tools::ReadFile`)
+- For `Cmd` tool, use the specific command names (e.g., `pwd`, `ls`)
+- When `available_tools` is not specified, all tools remain available (backward compatible)
+- Empty array (`available_tools: []`) means no tools for that step
+
+See the [available_tools_demo](examples/available_tools_demo/) for a complete example.
+
 #### ReadFile
 
 Reads the contents of a file from the filesystem.
@@ -811,14 +921,14 @@ tools:
   - Documentation:
       url: https://gitmcp.io/myorg/myrepo/docs
       env:
-        - "Authorization: Bearer {{env.API_TOKEN}}"
+        - "Authorization: Bearer {{ENV['API_TOKEN']}}"
 
   # MCP tools with stdio
   - GitHub:
       command: npx
       args: ["-y", "@modelcontextprotocol/server-github"]
       env:
-        GITHUB_PERSONAL_ACCESS_TOKEN: "{{env.GITHUB_TOKEN}}"
+        GITHUB_PERSONAL_ACCESS_TOKEN: "{{ENV['GITHUB_TOKEN']}}"
       only:
         - search_repositories
         - get_issue
@@ -847,7 +957,7 @@ Connect to local processes implementing the MCP protocol:
     command: docker
     args: ["run", "-i", "--rm", "ghcr.io/example/mcp-server"]
     env:
-      API_KEY: "{{env.API_KEY}}"
+      API_KEY: "{{ENV['API_KEY']}}"
 ```
 
 See the [MCP tools example](examples/mcp/) for complete documentation and more examples.

data/bin/roast

@@ -8,7 +8,7 @@
 # this file is here to facilitate running it.
 #
 
-ENV["BUNDLE_GEMFILE"] ||= File.expand_path("../.ruby-lsp/Gemfile", __dir__)
+ENV["BUNDLE_GEMFILE"] ||= File.expand_path("../Gemfile", __dir__)
 
 bundle_binstub = File.expand_path("bundle", __dir__)