← 返回 Skills 市场
457
总下载
0
收藏
0
当前安装
13
版本数
在 OpenClaw 中安装
/install modellix
功能描述
Integrate Modellix's unified API for AI image and video generation into applications. Use this skill whenever the user wants to generate images from text, cr...
使用说明 (SKILL.md)
Modellix Skill
Modellix is a Model-as-a-Service (MaaS) platform with async image/video generation APIs. The invariant flow is: submit task -> get task_id -> poll until success or failed.
Official Docs
- AI Onboarding (agent quick start): https://docs.modellix.ai/get-started.md
- API: https://docs.modellix.ai/ways-to-use/api.md
- CLI: https://docs.modellix.ai/ways-to-use/cli.md
- Full Models Docs Index: https://docs.modellix.ai/llms.txt
Execution Policy (CLI-first)
Always choose execution path in this order:
- Use CLI when
modellix-cliis available and authenticated. - Fall back to REST when CLI is unavailable, unsuitable, or missing capability.
- Prefer machine-readable outputs (
--json) in CLI flows.
Preflight and Deterministic Execution
Use bundled scripts before ad-hoc commands:
scripts/preflight.py- Validates CLI availability and API key presence.
- Returns recommended mode (
cliorrest).
scripts/invoke_and_poll.py- Executes CLI-first with REST fallback support.
- Handles exponential backoff polling and retryable submit errors.
- Emits normalized JSON result output.
Quick commands:
python scripts/preflight.py --json
python scripts/invoke_and_poll.py --model-slug bytedance/seedream-4.5-t2i --body '{"prompt":"A cinematic portrait of a fox in a misty forest at sunrise"}'
Core Workflow
1) Obtain API key
- Create key in Modellix Console
- Save immediately (shown once)
- Store as
MODELLIX_API_KEY
2) Select model
Read references/REFERENCE.md to find model docs and parameters.
3) Run invocation and poll
- Preferred:
scripts/invoke_and_poll.py - Manual CLI flow:
references/cli-playbook.md - Manual REST flow:
references/rest-playbook.md
4) Consume resources
Output media URLs are under result.resources. Persist assets promptly; results expire in 24 hours.
Progressive Reference Routing
Read only what the task needs:
references/cli-playbook.md- CLI install/auth/command flow and retry guidance
references/rest-playbook.md- REST endpoints, headers, status model, retry policy
references/capability-matrix.md- CLI command \x3C-> REST endpoint mapping and fallback rules
Bundled Assets
- Output schema:
assets/output/task-result.schema.json
Credential and Data Egress
- Required credential:
MODELLIX_API_KEY(this skill does not require any other secret). - Network egress: sends requests to
https://api.modellix.ai. - User payload handling: prompts and user-provided inputs (including media URLs or file-derived content) may be sent to Modellix endpoints during invocation.
- Result handling: generated resource URLs come from Modellix response payloads and should be downloaded before expiry (about 24 hours).
Error/Retry Policy
| Code | Action |
|---|---|
| 400 | Do not retry. Fix parameters or request body format. |
| 401 | Do not retry. Verify API key. |
| 402 | Do not retry. Insufficient balance. |
| 404 | Do not retry. Verify task_id or model-slug. |
| 429 | Retry with exponential backoff. |
| 500/503 | Retry with exponential backoff (max 3 times). |
Verification Checklist
- Preflight executed and mode selected (
cliorrest) - API key configured (
MODELLIX_API_KEYor CLI--api-key) - Model parameters verified against model doc from
references/REFERENCE.md - Task submit returns
task_idwith success code - Polling handles
pending,processing,success,failed - Retry behavior implemented for
429/500/503 - Result URLs persisted before 24-hour expiration
- REST fallback validated when CLI path is unavailable
安全使用建议
This skill appears to do what it says: it will send prompts, any provided media, and the API key to Modellix (https://api.modellix.ai) to create images/videos and may invoke a locally installed 'modellix-cli' if available. Before installing, ensure you trust the Modellix service and its CLI package (npm install -g modellix-cli), store your MODELLIX_API_KEY securely (do not paste it into untrusted places), and be aware generated assets expire (~24h) so persist anything you need. If you have privacy concerns about sending user data or files to an external API, do not enable the skill or use it only with non-sensitive inputs.
功能分析
Type: OpenClaw Skill
Name: modellix
Version: 1.0.12
The 'modellix' skill is a legitimate integration for the Modellix AI image and video generation platform. It provides scripts (scripts/invoke_and_poll.py and scripts/preflight.py) to interact with the service via its CLI or REST API (api.modellix.ai). The code follows security best practices by using subprocess.run without a shell for CLI calls and explicitly documenting data egress to the service's endpoints. No evidence of malicious intent, data exfiltration to unauthorized domains, or prompt injection attacks was found.
能力标签
能力评估
Purpose & Capability
Name/description match the implementation: scripts and docs implement CLI-first invocation and REST fallback against https://api.modellix.ai. No unrelated services, binaries, or extra credentials are requested.
Instruction Scope
Runtime instructions only use MODELLIX_API_KEY, optional body files, and optionally call the local 'modellix-cli'; they send user prompts/media to Modellix endpoints (explicitly documented). This external data egress is expected for a generation integration but is the primary privacy consideration.
Install Mechanism
No install spec; the skill is instruction-and-script based. Bundled scripts are plain Python and call either the CLI (if present) or the official API host. Nothing is downloaded from untrusted URLs or extracted to disk by the skill package itself.
Credentials
Only the single API key MODELLIX_API_KEY is required (declared in skill.json and SKILL.md). No other tokens, passwords, or config paths are requested. Access to the API key is justified by the described REST/CLI flows.
Persistence & Privilege
always is false and the skill contains no code that modifies other skills or system-wide settings. It does not request persistent elevated privileges.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install modellix - 安装完成后，直接呼叫该 Skill 的名称或使用
/modellix触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.12
- Simplified and clarified documentation for skill usage and configuration.
- Refined API key setup and management instructions for easier onboarding.
- Moved official links for CLI, API, and model docs to the top for quick reference.
- Condensed execution policy and workflow descriptions for improved readability.
- Maintained all core functionality and existing policies, focusing changes on guidance and structure.
v1.0.11
- Clarified that the default execution path is now the direct CLI command pair (`modellix-cli model invoke` and `modellix-cli task get`); bundled scripts like `invoke_and_poll.py` are termed optional helpers.
- Updated the "Core Workflow" and "Preflight and Deterministic Execution" sections to emphasize manual CLI as default, with automated scripts as alternatives.
- No functional logic or behavioral changes; documentation improvements only.
- Retained existing API key lifecycle and credential handling requirements.
v1.0.10
- Clarified CLI-first policy with explicit default CLI commands for task creation and result retrieval.
- Specified that deprecated CLI flags (e.g., --model-type) should not be used, and --help is only for troubleshooting.
- Added guidance for when the preflight check reports missing CLI: offer to install `modellix-cli` via npm with user consent before REST fallback.
- Improved details on the manual CLI command path and highlighted correct usage for core flows.
- No code or file changes detected; this was a documentation and guidance refinement release.
v1.0.9
Modellix Skill 1.0.9
- Added explicit primaryCredential and requiredEnv fields (MODELLIX_API_KEY) to the skill manifest for improved credential handling.
- Clarified credential and secret handling steps for both runtime environment and persistence scenarios.
- No functional or workflow changes; documentation updated to emphasize secure handling and environment configuration for API keys.
- No code or behavior changes detected.
v1.0.8
- Changes the API key lifecycle policy: credentials are now used in session scope by default and only persisted at the user’s explicit request.
- Removes support for agent-specific config-file persistence; persistent storage is limited to user-level environment variables (never system-wide or agent config).
- Clarifies discovery order: session environment variables take precedence, then user-level env vars.
- Explicitly forbids automatic credential persistence and writing to agent config or system-level env by default.
- Updates documentation to reflect the new explicit-consent flow for API key storage and session-only usage as standard.
v1.0.7
- Added a comprehensive API key lifecycle policy detailing discovery, request, persistence, and key rotation procedures.
- Enforced secret hygiene: API keys are never logged, echoed, or exposed; keys must be masked or avoided in output.
- Outlined priority for key persistence (environment variable first, agent config as fallback), with agent-specific handling guidance.
- Updated workflow to require preflight credential validation and blocking if a key is missing or invalid.
- Clarified agent config fallback for credential storage when env persistence is not possible.
- Expanded documentation to reflect secure credential management and error handling best practices.
v1.0.6
- Updated polling flow to handle the additional task status value: processing (now "pending", "processing", "success", "failed").
- Minor documentation formatting and command example fixes.
- No functional or file-level changes in this release.
v1.0.5
- Added compiled Python scripts to the repository: `scripts/__pycache__/clean_build_artifacts.cpython-314.pyc`, `preflight.cpython-314.pyc`, and `invoke_and_poll.cpython-314.pyc`
- Updated CLI invocation example in documentation to use `--model-slug bytedance/seedream-4.5-t2i` instead of provider/model-id pattern
v1.0.4
modellix 1.0.4
- Added CLI-first execution flow with fallback to REST, using bundled Python scripts for preflight checks and deterministic task invocation & polling.
- New references and scripts: CLI playbook, REST playbook, capability matrix, and utility scripts for validation and invocation.
- Introduced a JSON schema for task result output in assets/output.
- Enhanced documentation to outline workflow, retry/error policies, progressive reference routing, and a robust verification checklist.
- Clarified environment setup: only MODELLIX_API_KEY is required.
v1.0.3
- Added required environment variable definition for MODELLIX_API_KEY to configuration.
- Updated documentation to include env block specifying API key description and requirement.
- No changes to runtime logic or workflow.
v1.0.2
- Major cleanup: removed 21 files, including all .agents/skills/skill-creator source, scripts, evaluations, references, and auxiliary documentation.
- Removed top-level files: README.md and skills-lock.json.
- The skill now consists primarily of the core SKILL.md documentation for Modellix API integration.
v1.0.1
Modellix Skill 2.0 introduces a fully unified workflow, cleaner integration instructions, and robust error handling for AI image/video generation.
- Unified step-by-step workflow: from key creation to selecting models, submitting async tasks, polling for results, and handling errors.
- Clear instructions to use the model catalog (`references/REFERENCE.md`) to find and match the best model for each task.
- Expanded and unified error handling table, with retry logic guidance per HTTP status code.
- Concrete, ready-to-use code patterns for Python and Node.js REST integrations.
- Updated and broadened skill description and trigger conditions to cover all Modellix-supported task types and providers.
- All guidance now aligns tightly with the current official Modellix documentation and service practices.
v1.0.0
Modellix 1.0.0 – Initial Release
- Provides unified async API access to 100+ AI models for image and video generation across multiple providers.
- Supports text-to-image, text-to-video, image-to-image, image-to-video, and video editing tasks.
- Features a consistent task-based workflow: submit a task, receive a task ID, then poll for results.
- Includes guidance for error handling, rate limits, and best practices for async batch processing.
- Includes decision sections for async vs. streaming usage and retry logic for common error codes.
- Highlights key usage guidelines, pricing info, and integration checklists.
元数据
常见问题
Modellix 是什么？
Integrate Modellix's unified API for AI image and video generation into applications. Use this skill whenever the user wants to generate images from text, cr... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 457 次。
如何安装 Modellix？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install modellix」即可一键安装，无需额外配置。
Modellix 是免费的吗？
是的，Modellix 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Modellix 支持哪些平台？
Modellix 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Modellix？
由 modellix（@modellix）开发并维护，当前版本 v1.0.12。
推荐 Skills