@auto-engineer/model-factory 1.147.0 → 1.149.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (6) hide show
  1. package/.turbo/turbo-build.log +1 -1
  2. package/.turbo/turbo-test.log +3 -3
  3. package/.turbo/turbo-type-check.log +1 -1
  4. package/CHANGELOG.md +45 -0
  5. package/README.md +171 -0
  6. package/package.json +1 -1
package/.turbo/turbo-build.log CHANGED
@@ -1,5 +1,5 @@
1
1
 
2
- > @auto-engineer/model-factory@1.147.0 build /home/runner/work/auto-engineer/auto-engineer/packages/model-factory
2
+ > @auto-engineer/model-factory@1.149.0 build /home/runner/work/auto-engineer/auto-engineer/packages/model-factory
3
3
  > tsc && tsx ../../scripts/fix-esm-imports.ts
4
4
 
5
5
  Fixed ESM imports in dist/
package/.turbo/turbo-test.log CHANGED
@@ -1,5 +1,5 @@
1
1
 
2
- > @auto-engineer/model-factory@1.146.0 test /home/runner/work/auto-engineer/auto-engineer/packages/model-factory
2
+ > @auto-engineer/model-factory@1.148.0 test /home/runner/work/auto-engineer/auto-engineer/packages/model-factory
3
3
  > vitest run --reporter=dot
4
4
 
5
5
 
@@ -9,6 +9,6 @@
9
9
 
10
10
   Test Files  1 passed (1)
11
11
   Tests  5 passed (5)
12
-  Start at  19:11:30
13
-  Duration  1.52s (transform 295ms, setup 0ms, collect 234ms, tests 161ms, environment 0ms, prepare 374ms)
12
+  Start at  07:22:24
13
+  Duration  2.21s (transform 375ms, setup 0ms, collect 300ms, tests 154ms, environment 0ms, prepare 661ms)
14
14
 
package/.turbo/turbo-type-check.log CHANGED
@@ -1,4 +1,4 @@
1
1
 
2
- > @auto-engineer/model-factory@1.146.0 type-check /home/runner/work/auto-engineer/auto-engineer/packages/model-factory
2
+ > @auto-engineer/model-factory@1.148.0 type-check /home/runner/work/auto-engineer/auto-engineer/packages/model-factory
3
3
  > tsc --noEmit
4
4
 
package/CHANGELOG.md CHANGED
@@ -1,5 +1,50 @@
1
1
  # @auto-engineer/model-factory
2
2
 
3
+ ## 1.149.0
4
+
5
+ ### Minor Changes
6
+
7
+ - [`e1eebbd`](https://github.com/BeOnAuto/auto-engineer/commit/e1eebbdf4f209780e790094d2e6887c4fa809f98) Thanks [@github-actions[bot]](https://github.com/github-actions%5Bbot%5D)! - - **server-generator-apollo-emmett**: add Given state ref hints to state.ts.ejs
8
+ - **server-generator-apollo-emmett**: context-aware nonCommandField instructions
9
+ - **server-generator-apollo-emmett**: add state context instruction
10
+ - **server-generator-apollo-emmett**: extract shared template helpers
11
+ - **server-generator-apollo-emmett**: filter state refs from hasGivenEvents in decide.ts.ejs
12
+
13
+ ### Patch Changes
14
+
15
+ - [`d38c81e`](https://github.com/BeOnAuto/auto-engineer/commit/d38c81e7bb442a39626564cf4f6d8d55b60d0a38) Thanks [@SamHatoum](https://github.com/SamHatoum)! -
16
+
17
+ ## 1.148.0
18
+
19
+ ### Minor Changes
20
+
21
+ - [`d5ba3a0`](https://github.com/BeOnAuto/auto-engineer/commit/d5ba3a0e3fb0f6a9ad7a3a8b1815590ea77a5b42) Thanks [@rami-hatoum](https://github.com/rami-hatoum)! - - Added state context instruction to generated decide handlers, preventing unnecessary narrowing when Given steps contain only state references
22
+
23
+ - [`e0cdc4e`](https://github.com/BeOnAuto/auto-engineer/commit/e0cdc4e3363ad84d4bc49996a600ac75c97ccc38) Thanks [@rami-hatoum](https://github.com/rami-hatoum)! - - Added context-aware classification of non-command fields in generated decide.ts scaffolds, distinguishing between date-derived, state-derived, and not-yet-tested fields
24
+
25
+ - [`9195db7`](https://github.com/BeOnAuto/auto-engineer/commit/9195db78cb707d658866cee99a1c73d34fb4efde) Thanks [@rami-hatoum](https://github.com/rami-hatoum)! - - Extracted shared template helpers into a dedicated module for cleaner code generation
26
+
27
+ - [`abb6540`](https://github.com/BeOnAuto/auto-engineer/commit/abb6540db7196ed7935c8a8610695828f9035fc3) Thanks [@rami-hatoum](https://github.com/rami-hatoum)! - - Added status variant hints from Given state references to the state template, helping implementers create matching discriminated union variants
28
+
29
+ - [`9195db7`](https://github.com/BeOnAuto/auto-engineer/commit/9195db78cb707d658866cee99a1c73d34fb4efde) Thanks [@rami-hatoum](https://github.com/rami-hatoum)! - - Extracted shared template helper functions into a dedicated module for better code reuse across generators
30
+ - Simplified template specs by removing inline duplicate definitions in favor of the shared helpers
31
+
32
+ ### Patch Changes
33
+
34
+ - [`88fb1da`](https://github.com/BeOnAuto/auto-engineer/commit/88fb1da2b222de04dd4959d87657395ee960a6ce) Thanks [@github-actions[bot]](https://github.com/github-actions%5Bbot%5D)! - - **server-generator-apollo-emmett**: skip empty file plans in scaffold output
35
+ - **server-generator-apollo-emmett**: filter state refs from given() in decide.specs.ts.ejs
36
+ - **server-generator-apollo-emmett**: move CS Given states from events to states array
37
+ - **global**: version packages
38
+ - **server-generator-apollo-emmett**: mark G1+G2 ketchup plan complete
39
+
40
+ - [`4255f6d`](https://github.com/BeOnAuto/auto-engineer/commit/4255f6db0d128979e573244a615886482ce799b0) Thanks [@rami-hatoum](https://github.com/rami-hatoum)! - - Updated ketchup plan for state reference fix in decide template generator
41
+ - Marked generator bug fix milestones G1 and G2 as complete
42
+
43
+ - [`62f1ea3`](https://github.com/BeOnAuto/auto-engineer/commit/62f1ea3dd1b4275211574e3df9d9a6571ae9b27a) Thanks [@rami-hatoum](https://github.com/rami-hatoum)! - - Fixed scaffold generation to correctly distinguish between event and state references in decision handlers
44
+ - Prevented contradictory instructions from appearing in generated code when Given clauses contain only state references
45
+
46
+ - [`ba4f5c9`](https://github.com/BeOnAuto/auto-engineer/commit/ba4f5c9749fb1c15d444e78ca9a2689817f039cb) Thanks [@rami-hatoum](https://github.com/rami-hatoum)! - - Added implementation plan for decide.ts code generation fixes in the Apollo Emmett server generator
47
+
3
48
  ## 1.147.0
4
49
 
5
50
  ### Minor Changes
package/README.md ADDED
@@ -0,0 +1,171 @@
1
+ # @auto-engineer/model-factory
2
+
3
+ Creates a Vercel AI SDK `LanguageModel` from environment variables, connecting to any OpenAI-compatible provider.
4
+
5
+ ---
6
+
7
+ ## Purpose
8
+
9
+ Without `model-factory`, you would have to hard-code provider setup logic (provider name, base URL, API key, model ID) in every package that needs an LLM. This module centralises that configuration into four environment variables and returns a ready-to-use `LanguageModel` instance.
10
+
11
+ The design is deliberately minimal: one function, four env vars, no config files. Any OpenAI-compatible API gateway (LiteLLM, OpenRouter, a direct provider endpoint) works without code changes.
12
+
13
+ ## Key Concepts
14
+
15
+ - **OpenAI-compatible provider** -- The factory uses `@ai-sdk/openai-compatible` under the hood, so any provider that exposes an OpenAI-shaped chat completions API is supported.
16
+ - **Environment-driven configuration** -- All connection details come from `CUSTOM_PROVIDER_*` environment variables. No constructor arguments, no config objects.
17
+ - **Vercel AI SDK integration** -- The returned object satisfies the AI SDK `LanguageModel` interface and can be passed directly to `generateText`, `streamText`, and other AI SDK functions.
18
+
19
+ ---
20
+
21
+ ## Installation
22
+
23
+ ```bash
24
+ pnpm add @auto-engineer/model-factory
25
+ ```
26
+
27
+ Peer dependency:
28
+
29
+ ```bash
30
+ pnpm add ai
31
+ ```
32
+
33
+ ---
34
+
35
+ ## Quick Start
36
+
37
+ Set the required environment variables:
38
+
39
+ ```bash
40
+ export CUSTOM_PROVIDER_NAME="litellm"
41
+ export CUSTOM_PROVIDER_BASE_URL="https://gateway.example.com/v1"
42
+ export CUSTOM_PROVIDER_API_KEY="sk-..."
43
+ export CUSTOM_PROVIDER_DEFAULT_MODEL="xai/grok-code-fast-1"
44
+ ```
45
+
46
+ Then create and use a model:
47
+
48
+ ```typescript
49
+ import { createModelFromEnv } from '@auto-engineer/model-factory';
50
+ import { generateText } from 'ai';
51
+
52
+ const model = createModelFromEnv();
53
+
54
+ const { text } = await generateText({
55
+ model,
56
+ prompt: 'Explain monads in one sentence.',
57
+ });
58
+ ```
59
+
60
+ ---
61
+
62
+ ## How-to Guides
63
+
64
+ ### Connect to a different provider
65
+
66
+ Change the environment variables -- no code changes required:
67
+
68
+ ```bash
69
+ # OpenRouter
70
+ export CUSTOM_PROVIDER_NAME="openrouter"
71
+ export CUSTOM_PROVIDER_BASE_URL="https://openrouter.ai/api/v1"
72
+ export CUSTOM_PROVIDER_API_KEY="sk-or-..."
73
+ export CUSTOM_PROVIDER_DEFAULT_MODEL="anthropic/claude-sonnet-4-20250514"
74
+ ```
75
+
76
+ ### Use in tests
77
+
78
+ Mock the module or set test-specific env vars before importing:
79
+
80
+ ```typescript
81
+ process.env.CUSTOM_PROVIDER_NAME = 'test-provider';
82
+ process.env.CUSTOM_PROVIDER_BASE_URL = 'http://localhost:4000/v1';
83
+ process.env.CUSTOM_PROVIDER_API_KEY = 'test-key';
84
+ process.env.CUSTOM_PROVIDER_DEFAULT_MODEL = 'test-model';
85
+
86
+ const { createModelFromEnv } = await import('@auto-engineer/model-factory');
87
+ const model = createModelFromEnv();
88
+ ```
89
+
90
+ ---
91
+
92
+ ## API Reference
93
+
94
+ ### Package Exports
95
+
96
+ | Export | Kind | Description |
97
+ |---|---|---|
98
+ | `createModelFromEnv` | Function | Creates a `LanguageModel` from environment variables |
99
+ | `ChatMessage` | Interface | A single chat message with role and content |
100
+ | `ChatTurn` | Interface | A prompt/response pair |
101
+
102
+ ### Functions
103
+
104
+ #### `createModelFromEnv()`
105
+
106
+ Creates a Vercel AI SDK `LanguageModel` by reading four required environment variables.
107
+
108
+ ```typescript
109
+ function createModelFromEnv(): LanguageModel
110
+ ```
111
+
112
+ **Environment variables (all required):**
113
+
114
+ | Variable | Description |
115
+ |---|---|
116
+ | `CUSTOM_PROVIDER_NAME` | Identifier for the provider (e.g. `"litellm"`) |
117
+ | `CUSTOM_PROVIDER_BASE_URL` | Base URL of the OpenAI-compatible API |
118
+ | `CUSTOM_PROVIDER_API_KEY` | API key for authentication |
119
+ | `CUSTOM_PROVIDER_DEFAULT_MODEL` | Model identifier to use (e.g. `"xai/grok-code-fast-1"`) |
120
+
121
+ **Throws:** `Error` if any of the four environment variables is missing or empty.
122
+
123
+ **Returns:** A `LanguageModel` instance backed by the configured provider.
124
+
125
+ ### Interfaces
126
+
127
+ #### `ChatMessage`
128
+
129
+ ```typescript
130
+ interface ChatMessage {
131
+ role: 'user' | 'assistant';
132
+ content: string;
133
+ }
134
+ ```
135
+
136
+ #### `ChatTurn`
137
+
138
+ ```typescript
139
+ interface ChatTurn {
140
+ prompt: ChatMessage;
141
+ response: ChatMessage;
142
+ }
143
+ ```
144
+
145
+ ---
146
+
147
+ ## Architecture
148
+
149
+ ```
150
+ src/
151
+ index.ts # createModelFromEnv, ChatMessage, ChatTurn
152
+ index.specs.ts # Unit tests
153
+ ```
154
+
155
+ ### Dependencies
156
+
157
+ | Package | Purpose |
158
+ |---|---|
159
+ | `@ai-sdk/openai-compatible` | Creates the underlying provider client |
160
+ | `ai` (peer) | Provides the `LanguageModel` type |
161
+
162
+ ### Unused direct dependencies
163
+
164
+ The following packages are listed in `dependencies` but are not currently imported in source code. They may be intended for future provider-specific factories:
165
+
166
+ | Package |
167
+ |---|
168
+ | `@ai-sdk/anthropic` |
169
+ | `@ai-sdk/google` |
170
+ | `@ai-sdk/openai` |
171
+ | `@ai-sdk/xai` |
package/package.json CHANGED
@@ -22,7 +22,7 @@
22
22
  "publishConfig": {
23
23
  "access": "public"
24
24
  },
25
- "version": "1.147.0",
25
+ "version": "1.149.0",
26
26
  "scripts": {
27
27
  "build": "tsc && tsx ../../scripts/fix-esm-imports.ts",
28
28
  "test": "vitest run --reporter=dot",