Add MiniMax engine support - #175
Open
octo-patch wants to merge 1 commit into
Open
Conversation
|
@octo-patch is attempting to deploy a commit to the Goshen Labs Team on Vercel. A member of the Team first needs to authorize it. |
📄 Knowledge review✏️ Suggested updates2 page suggestions need review.
📝 AI Engine Addition Process@@ -1,10 +1,10 @@
## Architecture and Extension Points
AI engines in Ralphy are implemented as classes extending a shared `BaseAIEngine` abstract class. Each engine defines its name, CLI command, and execution logic. Engines are registered in `cli/src/engines/index.ts` and instantiated via the `createEngine` function, which maps engine names to their respective classes. This modular approach ensures consistent integration and simplifies extension for new engines ([source](https://github.com/michaelshimeles/ralphy/blob/fc2df589969b5fe16d31eccb4e7ff91314e31776/cli/src/engines/index.ts#L3-L56)).
-Supported engines include Claude Code, OpenCode, Cursor, Codex, Qwen-Code, Factory Droid, GitHub Copilot, Trae Agent, Gemini CLI, and Ollama (via Claude Code CLI). Each engine follows the same integration pattern, allowing for consistent behavior and easy extensibility.
+Supported engines include Claude Code, OpenCode, Cursor, Codex, Qwen-Code, Factory Droid, GitHub Copilot, Trae Agent, Gemini CLI, MiniMax, and Ollama (via Claude Code CLI). Each engine follows the same integration pattern, allowing for consistent behavior and easy extensibility.
## Adding Command-Line Flags
-To add a new engine, define a unique command-line flag in the CLI argument parser. Ralphy uses the `commander` library in its TypeScript CLI to declare flags such as `--droid` for Factory Droid, `--qwen` for Qwen-Code, `--trae` for Trae Agent, `--gemini` for Gemini CLI, and `--ollama` for Ollama. Update the argument parsing logic to set the engine name when the flag is present ([source](https://github.com/michaelshimeles/ralphy/blob/fc2df589969b5fe16d31eccb4e7ff91314e31776/cli/src/cli/args.ts#L26-L145)).
+To add a new engine, define a unique command-line flag in the CLI argument parser. Ralphy uses the `commander` library in its TypeScript CLI to declare flags such as `--droid` for Factory Droid, `--qwen` for Qwen-Code, `--trae` for Trae Agent, `--gemini` for Gemini CLI, `--minimax` for MiniMax, and `--ollama` for Ollama. Update the argument parsing logic to set the engine name when the flag is present ([source](https://github.com/michaelshimeles/ralphy/blob/fc2df589969b5fe16d31eccb4e7ff91314e31776/cli/src/cli/args.ts#L26-L145)).
Example:
```typescript
@@ -13,6 +13,7 @@
.option("--qwen", "Use Qwen-Code")
.option("--trae", "Use Trae Agent")
.option("--gemini", "Use Gemini CLI")
+ .option("--minimax", "Use MiniMax through Claude Code")
.option("--ollama", "Use Ollama (local models via Claude Code)");
```
Argument parsing:
@@ -26,6 +27,7 @@
else if (opts.droid) aiEngine = "droid";
else if (opts.trae) aiEngine = "trae";
else if (opts.gemini) aiEngine = "gemini";
+else if (opts.minimax) aiEngine = "minimax";
else if (opts.ollama) aiEngine = "ollama";
```
📝 AI Engine Integration@@ -6,7 +6,7 @@
**Engine-Specific Arguments:** You can pass arbitrary arguments to any engine using the `--` separator. Everything after `--` is forwarded directly to the engine CLI. See the 'Engine-Specific Arguments' section for details and examples.
**Requirements:**
-- AI CLI: [Claude Code](https://github.com/anthropics/claude-code), [OpenCode](https://opencode.ai/docs/), [Cursor](https://cursor.com), Codex, Qwen-Code, [Factory Droid](https://docs.factory.ai/cli/getting-started/quickstart), [GitHub Copilot](https://docs.github.com/en/copilot/how-tos/use-copilot-agents/use-copilot-cli), [Gemini CLI](https://github.com/google-gemini/gemini-cli), or [Ollama](https://ollama.com) (requires Claude Code CLI)
+- AI CLI: [Claude Code](https://github.com/anthropics/claude-code), [OpenCode](https://opencode.ai/docs/), [Cursor](https://cursor.com), Codex, Qwen-Code, [Factory Droid](https://docs.factory.ai/cli/getting-started/quickstart), [GitHub Copilot](https://docs.github.com/en/copilot/how-tos/use-copilot-agents/use-copilot-cli), [Gemini CLI](https://github.com/google-gemini/gemini-cli), MiniMax (requires Claude Code CLI), or [Ollama](https://ollama.com) (requires Claude Code CLI)
- **npm version (`ralphy-cli`)**: Node.js 18+ or Bun
Each engine requires its CLI tool installed and available in the system PATH.
@@ -23,6 +23,7 @@
| Trae Agent | `trae` | `--trae` | `--print --force --output-format stream-json` | stream-json | tokens, duration (ms) | Model override via `--model`. Trae CLI must be in PATH. |
| GitHub Copilot| `copilot` | `--copilot` | `--yolo` | plain text | tokens (parsed from output) | Requires Copilot CLI. **Prompts are passed via temp files to preserve markdown. --yolo is always used for non-interactive mode. Only authentication errors with known output formats are detected and surfaced. Token usage is parsed and reported. Silent failures and infinite retry loops are fixed.** |
| Gemini | `gemini` | `--gemini` | `--output-format stream-json --yolo` | stream-json | tokens + cost | Model override via `--model`. Failed commands return error messages with exit codes. |
+| MiniMax | `claude` | `--minimax` | `--dangerously-skip-permissions` | stream-json | tokens + cost | Uses Claude Code transport with MiniMax endpoint. Requires `MINIMAX_API_KEY` environment variable. Default model: `MiniMax-M3`. |
| Ollama | `claude` | `--ollama` | `--dangerously-skip-permissions` (Ollama env vars) | stream-json | tokens + cost | Runs local models via Claude Code CLI. Requires Ollama running locally and Claude Code CLI installed. Recommended models: `qwen3-coder`, `glm-4.7`, `gpt-oss:20b`, `gpt-oss:120b`. |
## Engine Integration Details
@@ -89,6 +90,15 @@
Example:
```bash
ralphy --gemini --model gemini-pro "implement feature"
+```
+
+### MiniMax
+Integrated as `MiniMaxEngine`. Uses the `claude` CLI with the MiniMax endpoint. Model override via `--model <name>`. Default model is `MiniMax-M3`. Requires `MINIMAX_API_KEY` environment variable for authentication. Token usage and cost are parsed from output. Failed commands return error messages with exit codes.
+
+Example:
+```bash
+export MINIMAX_API_KEY="..."
+ralphy --minimax "add feature"
```
### GitHub Copilot
@@ -166,6 +176,7 @@
- **Claude Code**: Authentication/connectivity issues may occur depending on environment; not considered a bug in Ralphy.
- **Codex**: No token reporting; uses temp files for output.
- **GitHub Copilot**: Requires Copilot CLI installed and available in PATH. Prompts are passed via temp files. --yolo is always used for non-interactive mode. Only authentication errors with known output formats are detected and surfaced. Token usage is parsed and reported. Infinite retry loops on fatal errors are fixed.
+- **MiniMax**: Requires `MINIMAX_API_KEY` environment variable and Claude Code CLI installed. Uses Claude Code transport with MiniMax endpoint.
- **Ollama**: Requires [Ollama](https://ollama.com) running locally and Claude Code CLI installed and available in PATH. Only models with at least 64k context window are supported. If either dependency is missing, tasks will fail with a clear error message.
- **General**: Each engine requires its CLI tool installed and available in the system PATH.
@@ -180,6 +191,7 @@
ralphy --trae "implement feature" # Trae Agent
ralphy --copilot "add feature" # GitHub Copilot
ralphy --gemini "summarize document" # Gemini CLI
+ralphy --minimax "add feature" # MiniMax (through Claude Code)
ralphy --ollama "add feature" # Ollama (local models via Claude Code CLI)
ralphy --opencode --model opencode/glm-4.7-free "custom model"
|
Contributor
Greptile SummaryThis change adds MiniMax as a Claude Code-backed engine, including CLI selection, endpoint configuration, model defaults, tests, and documentation. Validation reproduced two issues:
Confidence Score: 4/5
|
![greptile-apps[bot]](https://avatars.githubusercontent.com/in/867647?s=60&v=4){kind=small}
|
|
||
| protected getEnvironment(options?: EngineOptions): Record<string, string> { | ||
| const model = this.getModel(options); | ||
| const authToken = process.env.MINIMAX_API_KEY ?? process.env.ANTHROPIC_AUTH_TOKEN; |
Contributor
There was a problem hiding this comment.
When --minimax runs without MINIMAX_API_KEY while ANTHROPIC_AUTH_TOKEN is set, the engine passes the Anthropic bearer token to a Claude subprocess configured for https://api.minimax.io/anthropic. This exposes an unrelated provider credential to MiniMax and will generally fail MiniMax authentication. Require a MiniMax-scoped credential instead of falling back to ANTHROPIC_AUTH_TOKEN.
Artifacts
Credential-boundary runtime probe script
- A temporary fake Claude executable records the environment received from a real MiniMax engine invocation, proving the exact child-process boundary exercised.
MiniMax credential-boundary probe output with Anthropic fallback token
- The executed probe exited successfully and recorded the MiniMax endpoint together with the injected Anthropic sentinel token, confirming the credential-boundary defect.
Prompt To Fix With AI
This is a comment left during a code review.
Path: cli/src/engines/minimax.ts
Line: 19
Comment:
**Cross-provider credential fallback**
When `--minimax` runs without `MINIMAX_API_KEY` while `ANTHROPIC_AUTH_TOKEN` is set, the engine passes the Anthropic bearer token to a Claude subprocess configured for `https://api.minimax.io/anthropic`. This exposes an unrelated provider credential to MiniMax and will generally fail MiniMax authentication. Require a MiniMax-scoped credential instead of falling back to `ANTHROPIC_AUTH_TOKEN`.
---
For each issue above, determine whether it is valid and should be fixed. If so, fix it directly.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Reason: Add first-class MiniMax engine support through the Claude Code transport.
Changes
--minimaxengine selection withMiniMax-M3as the default modelMINIMAX_API_KEYto bearer authenticationChecks
bun test src/engines/minimax.test.tsbun testbun run buildbunx biome check src/cli/args.ts src/engines/claude.ts src/engines/index.ts src/engines/types.ts src/engines/minimax.ts src/engines/minimax.test.tsgit diff --check origin/main...HEADbun run checkcontinues to report existing lint errors in execution and telemetry files outside this change.