mcpField
M
minimax-runner

minimax-runner

github.com/robsonrungAgent workflows247 installsUpdated 1d ago
Repository

About minimax-runner

minimax-runner is a skill for running prompts against Minimax models through the shared cline-runner implementation. It adds a headless Minimax seat for councils and scripted workflows, without falling back to another provider when execution fails.

Key features

  1. 01Runs prompts with Minimax models
  2. 02Uses the shared cline-runner wrapper
  3. 03Supports prompt files and role selection
  4. 04Emits stream output and a shared JSON envelope
  5. 05Keeps failed Minimax runs unavailable rather than substituting

Use cases

  1. 01Adding a Minimax seat to a model council
  2. 02Running scripted prompt workflows headlessly
  3. 03Stress-testing an implementation idea
  4. 04Comparing runner outputs in a consistent envelope

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/robsonrung/rar-skills --skill minimax-runner

After installing, type/minimax-runnerin Claude Code to confirm it loaded.

Bundled files

References and scripts installed along with SKILL.md.

BACKLOG.md
agents/
openai.yaml
scripts/
run_minimax.py

Details

From the project README

Minimax Runner

Execute prompts against Minimax models through the shared cline-runner implementation. This skill adds a Minimax-backed seat for councils and scripted workflows while keeping execution headless and stream friendly.

Default Model

  • minimax/minimax-m2.7

Pass --model if you want to target another Minimax model exposed by the selected Cline provider.

Prerequisites

  • cline installed and in PATH
  • A Cline provider authenticated through cline auth that can resolve minimax/minimax-m2.7

Security Model

This skill delegates to cline-runner, so it has the Cline execution and data sharing model. Prompt text, prompt files, session files, metadata, and any files Cline reads during the run may be sent to the selected Minimax provider. Analysis roles default to native --auto-approve false; pass --allow-write to opt out.

Shared Wrapper Reference

Supported options, roles, the --json output envelope key contract, return codes, and gotchas are identical to the shared wrapper — read the cline-runner skill's SKILL.md (../cline-runner/SKILL.md) when you need flag or envelope details. The envelope is produced by cline-runner/scripts/run_cline.py.

Usage

python3 .agents/skills/minimax-runner/scripts/run_minimax.py "your prompt here"

Examples

python3 .agents/skills/minimax-runner/scripts/run_minimax.py "Stress-test this implementation idea"
python3 .agents/skills/minimax-runner/scripts/run_minimax.py --prompt-file /tmp/review.md --role challenger
python3 .agents/skills/minimax-runner/scripts/run_minimax.py "Return JSON only" --output-format stream-json --json

Behavior

  1. Delegates to the shared cline-runner implementation with runner identity set to minimax.
  2. Uses the Cline NDJSON stream as the default output format.
  3. Never falls back to another provider. If the run fails (non-zero return_code or auth_ok=false in the envelope), treat the Minimax seat as blocked and report it unavailable so councils can account for the missing seat — never substitute another provider.
  4. Preserves the shared wrapper envelope so councils can compare Minimax output with other runners consistently.

Comments

Markdown supported