@inference-gateway/adl-cli 0.66.3 → 0.67.0

package/README.md

@@ -63,7 +63,7 @@ The ADL CLI helps you build enterprise-ready A2A agents quickly by generating co
 - 🔧 **CI/CD Generation** - Automatic GitHub Actions workflows with semantic-release CD pipelines
 - 🏗️ **Sandbox Environments** - Flox and DevContainer support for isolated development
 - 🎣 **Post-Generation Hooks** - Customize build, format, and test commands after generation
-- 🤖 **Multi-Provider AI** - OpenAI, Anthropic, Google, Groq, Mistral, DeepSeek, Cohere, Cloudflare, Moonshot, Ollama, Ollama Cloud, and Nvidia support
+- 🤖 **Multi-Provider AI** - OpenAI, Anthropic, Google, Groq, Mistral, DeepSeek, Cohere, Cloudflare, Moonshot, Ollama, Ollama Cloud, Nvidia, and MiniMax support
 - 📁 **Artifacts Support** - Integrated filesystem and MinIO object storage for artifact management
 - 📡 **OpenTelemetry Instrumentation** - Opt-in tracing and metrics via `spec.telemetry.enabled` (Go and TypeScript)
 - 🔌 **MCP Client** - Connect to MCP servers via `spec.agent.mcp` for `mcp_list_tools` / `mcp_call_tool` (Go)
@@ -187,15 +187,17 @@ adl init my-weather-agent
 adl generate --file agent.yaml --output ./test-my-agent
 ```
 
+The wizard defaults to `typescript` and adds no tools unless you answer yes to "Add tools to your agent". The Go example below assumes you picked `go` (for example with `--language go`) and added a `get_weather` tool.
+
 ### 2. Implement Your Business Logic
 
-The generated project includes TODO placeholders for your implementations:
+The generated project includes TODO placeholders for your implementations - here `tools/get_weather.go`:
 
 ```go
-// TODO: Implement weather API logic
-func GetWeatherTool(ctx context.Context, args map[string]any) (string, error) {
+// GetWeatherHandler handles the get_weather tool execution
+func (t *GetWeatherTool) GetWeatherHandler(ctx context.Context, args map[string]any) (string, error) {
+    // TODO: Implement get_weather logic
     city := args["city"].(string)
-    // TODO: Replace with actual weather API call
     return fmt.Sprintf(`{"city": "%s", "temp": "22°C"}`, city), nil
 }
 ```
@@ -203,7 +205,7 @@ func GetWeatherTool(ctx context.Context, args map[string]any) (string, error) {
 ### 3. Build and Run
 
 ```bash
-cd test-weather-agent
+cd test-my-agent
 task build
 task run
 ```
@@ -264,7 +266,6 @@ The init command supports extensive configuration options:
 
 - `--streaming` - Enable streaming responses
 - `--notifications` - Enable push notifications
-- `--history` - Enable state transition history
 
 **Server Configuration:**
 
@@ -278,7 +279,7 @@ The init command supports extensive configuration options:
 **Go Options:**
 
 - `--go-module` - Go module path (e.g., `github.com/user/project`)
-- `--go-version` - Go version (e.g., `1.26.7`)
+- `--go-version` - Go version (e.g., `1.26.8`)
 
 **Rust Options:**
 
@@ -500,7 +501,7 @@ spec:
   language:
     go:
       module: "github.com/example/weather-agent"
-      version: "1.26.7"
+      version: "1.26.8"
   acronyms: # Optional: Custom acronyms for better code generation
     - api
     - json
@@ -515,7 +516,7 @@ The complete ADL schema includes:
 - **capabilities**: Streaming, notifications, state history
 - **config**: Structured configuration sections with environment variable mapping
 - **services**: Service services with interfaces, factories, and type definitions
-- **agent**: AI provider configuration (OpenAI, Anthropic, Google, Groq, Mistral, DeepSeek, Cohere, Cloudflare, Moonshot, Ollama, Ollama Cloud, Nvidia)
+- **agent**: AI provider configuration (OpenAI, Anthropic, Google, Groq, Mistral, DeepSeek, Cohere, Cloudflare, Moonshot, Ollama, Ollama Cloud, Nvidia, MiniMax)
 - **tools**: Function-call definitions with JSON schemas, validation, and service injection support
 - **skills**: Markdown playbooks (id + optional `bare`, version, source) pulled from the skills registry, fetched as a full directory from a GitHub repo (shorthand or URL), or scaffolded locally into `.agents/skills/<id>/`; advertised on the agent card, and summarised at runtime in an `AVAILABLE SKILLS:` block appended to the system prompt (frontmatter only - the model reads each `SKILL.md` body on demand)
 - **server**: HTTP server configuration with authentication support
@@ -635,7 +636,7 @@ spec:
   language:
     go:
       module: "github.com/company/advanced-agent"
-      version: "1.26.7"
+      version: "1.26.8"
   scm:
     provider: github
     url: "https://github.com/company/advanced-agent"
@@ -707,7 +708,7 @@ spec:
   language:
     go:
       module: github.com/example/agent
-      version: "1.26.7"
+      version: "1.26.8"
       vendor:
         deps:
           - github.com/google/uuid@v1.6.0
@@ -803,7 +804,7 @@ Worked example per backend:
 # Flox: pin deno + kubectl + terraform alongside the Go toolchain
 spec:
   language:
-    go: { module: github.com/example/agent, version: "1.26.7" }
+    go: { module: github.com/example/agent, version: "1.26.8" }
   development:
     sandbox:
       flox:
@@ -818,7 +819,7 @@ spec:
 # Devcontainer: same deps, rendered as an apt-packages feature
 spec:
   language:
-    go: { module: github.com/example/agent, version: "1.26.7" }
+    go: { module: github.com/example/agent, version: "1.26.8" }
   development:
     sandbox:
       devcontainer:
@@ -1109,7 +1110,7 @@ type CacheConfig struct {
 
 ### Config Subsection Injection
 
-In addition to injecting entire configuration objects, you can inject specific config subsections directly into skills using dotted notation. This provides type-safe access to focused configuration scopes.
+In addition to injecting entire configuration objects, you can inject specific config subsections directly into tools (`spec.tools[].inject`) using dotted notation. This provides type-safe access to focused configuration scopes.
 
 **Example ADL Configuration:**
 
@@ -1199,11 +1200,11 @@ toolBox.AddTool(exportReportTool)
 
 **Benefits of Config Subsection Injection:**
 
-- **Scoped Access**: Skills only receive the configuration they need, following principle of least privilege
+- **Scoped Access**: Tools only receive the configuration they need, following principle of least privilege
 - **Type Safety**: Compile-time validation ensures config fields exist
-- **Clear Dependencies**: Explicit declaration of which config sections each skill requires
+- **Clear Dependencies**: Explicit declaration of which config sections each tool requires
 - **Easier Testing**: Mock specific config subsections without full config object
-- **Better Separation**: Skills don't have access to unrelated configuration
+- **Better Separation**: Tools don't have access to unrelated configuration
 - **Auto-Validation**: ADL CLI validates that injected config sections exist in `spec.config`
 
 **Injection Patterns:**
@@ -1225,7 +1226,7 @@ The service injection system generates:
 2. **Type-Safe Configuration**: Structured config with environment variable mapping
 3. **Service Interfaces**: Custom service packages with interface definitions
 4. **Factory Functions**: Constructor functions that receive logger and configuration
-5. **Automatic Registration**: Services are automatically wired into skills
+5. **Automatic Registration**: Services are automatically wired into tools
 6. **File Protection**: Generated service files are automatically added to `.adl-ignore`
 
 ### Generated Structure
@@ -1278,9 +1279,9 @@ func NewCalendarService(logger *zap.Logger, cfg *config.Config) (CalendarService
 }
 ```
 
-### Skill Integration
+### Tool Integration
 
-Skills automatically receive injected services as constructor parameters:
+Tools automatically receive injected services as constructor parameters:
 
 **Example `tools/create_event.go`:**
 
@@ -1878,7 +1879,7 @@ When `github_app: true` is set, the generated CD pipeline will use GitHub App au
 
 ### AI Provider Support
 
-The ADL CLI supports multiple AI providers including OpenAI, Anthropic, Google AI, Groq, Mistral, DeepSeek, Cohere, Cloudflare, Moonshot, Ollama (for local LLMs), Ollama Cloud, and Nvidia. Each provider requires appropriate API keys to be configured as environment variables. See the ADL examples above for configuration details.
+The ADL CLI supports multiple AI providers including OpenAI, Anthropic, Google AI, Groq, Mistral, DeepSeek, Cohere, Cloudflare, Moonshot, Ollama (for local LLMs), Ollama Cloud, Nvidia, and MiniMax. Each provider requires appropriate API keys to be configured as environment variables. See the ADL examples above for configuration details.
 
 ## Artifacts Support
 
@@ -1890,7 +1891,7 @@ spec:
     enabled: true
 ```
 
-Configure storage via environment variables (see generated README for A2A*ARTIFACT*\* variables). Supports both filesystem and MinIO/S3 storage backends.
+Configure storage via environment variables (see the generated `CONFIGURATIONS.md` for the `A2A_ARTIFACTS_*` variables, such as `A2A_ARTIFACTS_ENABLED` and `A2A_ARTIFACTS_STORAGE_PROVIDER`). Supports both filesystem and MinIO/S3 storage backends.
 
 **Examples:**
 
@@ -1974,17 +1975,15 @@ spec:
 
 When `issue_templates: true` is set, the following templates are generated in `.github/ISSUE_TEMPLATE/`:
 
-- **`bug_report.md`** - Structured bug reporting with severity levels, reproduction steps, and environment details
-- **`feature_request.md`** - Feature proposals with use case descriptions and acceptance criteria
-- **`refactor_request.md`** - Code improvement requests with motivation and impact analysis
+- **`bug_report.md`** - Summary, Steps to Reproduce, Expected Behavior and Actual Behavior
+- **`feature_request.md`** - Summary plus an Acceptance Criteria checklist
+- **`refactor_request.md`** - Summary plus an Acceptance Criteria checklist
 
 **Issue Template Features:**
 
-- **Agent Context** - Templates include agent name and version from your ADL metadata
 - **Structured Sections** - Consistent formatting for better issue triage and tracking
-- **GitHub Integration** - Automatic labels and assignees configured in frontmatter
-- **Severity Levels** - Priority classification for bug reports (critical, high, medium, low)
-- **Environment Info** - Sections for capturing logs, system details, and configurations
+- **Title Prefixes** - Frontmatter prefills titles with `[BUG]`, `[FEATURE]` and `[TASK] Refactor` (each followed by a trailing space)
+- **GitHub Integration** - Frontmatter sets labels (`bug`, `enhancement`, `refactor`) and issue types (`bug`, `feature`, `task`); `assignees` is left empty for you to fill in
 
 ## Dependabot Configuration
 
@@ -2046,7 +2045,7 @@ adl generate --file examples/cloudrun-agent.yaml --output ./cloudrun-enterprise
 - `cloudrun-agent.yaml` - Cloud Run deployment with Google Container Registry
 - `cloudrun-ghcr-agent.yaml` - Cloud Run deployment with GitHub Container Registry
 
-See [`examples/README.md`](examples/README.md) for the full catalog of all 18 example manifests.
+See [`examples/README.md`](examples/README.md) for the full catalog of all 22 example manifests.
 
 ## Template System & Architecture
 
@@ -2193,7 +2192,7 @@ You can control which additional files are generated or updated by editing the `
 # .adl-ignore
 # Skip Docker-related files if you have custom containerization
 Dockerfile
-docker-compose.yml
+docker-compose.yaml
 
 # Skip Kubernetes manifests if you use different deployment tools
 k8s/
@@ -2216,7 +2215,7 @@ Taskfile.yml
 
 ### Common Use Cases
 
-- **Custom Deployment**: Skip `Dockerfile`, `k8s/`, `docker-compose.yml`
+- **Custom Deployment**: Skip `Dockerfile`, `k8s/`, `docker-compose.yaml`
 - **Custom Build**: Skip `Taskfile.yml`, `Makefile`
 - **Custom Auth**: Skip `auth.go`, `middleware.go`
 - **Custom Documentation**: Skip `README.md`
@@ -2225,7 +2224,7 @@ Taskfile.yml
 
 ### Prerequisites
 
-- Go 1.26.7+
+- Go 1.26.8+
 - [Task](https://taskfile.dev/) (optional, for using Taskfile commands)
 
 ### Building from Source
@@ -2290,21 +2289,23 @@ spec:
   language:
     go:
       module: "github.com/company/my-agent"
-      version: "1.26.7"
+      version: "1.26.8"
   acronyms: ["n8n", "xml", "mqtt", "iot", "uuid"]
 ```
 
 ### Generated Code Examples
 
+Tool identifiers are generated as `<Name>Tool` (see `internal/templates/languages/go/tool.go.tmpl`). `xml` is already a default acronym, so only words outside the defaults change between the two cases below.
+
 **Without custom acronyms:**
 
-- `get_n8n_docs` → `GetN8nDocsSkill`
-- `process_xml_data` → `ProcessXmlDataSkill`
+- `get_n8n_docs` → `GetN8nDocsTool`
+- `publish_mqtt_event` → `PublishMqttEventTool`
 
-**With custom acronyms:**
+**With custom acronyms (`["n8n", "mqtt"]`):**
 
-- `get_n8n_docs` → `GetN8NDocsSkill`
-- `process_xml_data` → `ProcessXMLDataSkill`
+- `get_n8n_docs` → `GetN8NDocsTool`
+- `publish_mqtt_event` → `PublishMQTTEventTool`
 
 ### Default Acronyms
 
@@ -2313,6 +2314,7 @@ The following acronyms are recognized by default:
 - **Common**: id, api, url, uri, json, xml, sql, html, css, js, ui, uuid
 - **Network**: http, https, tcp, udp, ip, dns, tls, ssl
 - **Tech**: cpu, gpu, ram, io, os, db
+- **Units**: mb, gb, kb
 
 Your custom acronyms extend these defaults and take precedence over them.
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@inference-gateway/adl-cli",
-  "version": "0.66.3",
+  "version": "0.67.0",
   "description": "Scaffold complete A2A (Agent-to-Agent) projects from YAML Agent Definition Language (ADL) manifests. Thin npm wrapper around the native adl binary.",
   "bin": {
     "adl": "bin/run.js"