← 返回 Skills 市场

Filtrix Image Gen

Name: Filtrix Image Gen
Author: lumenclaw-cloud

作者 lumenclaw-cloud · GitHub ↗ · v1.0.2 · MIT-0

cross-platform ⚠ suspicious

518

总下载

当前安装

版本数

在 OpenClaw 中安装

/install filtrix-image-gen

功能描述

Generate and edit images through Filtrix Remote MCP. Use when users ask to create images or refine existing ones. Supports gpt-image-1, nano-banana, and nano...

使用说明 (SKILL.md)

Filtrix Image Gen (MCP)

This skill is MCP-only.

Endpoint: https://mcp.filtrix.ai/mcp
Auth: Authorization: Bearer \x3CFILTRIX_MCP_API_KEY>
Primary tools:
- generate_image_text
- edit_image_text
- get_account_credits

Available MCP tools:

get_account_credits
generate_image_text
edit_image_text

Setup

Required:

FILTRIX_MCP_API_KEY

Optional:

FILTRIX_MCP_URL (default: https://mcp.filtrix.ai/mcp)

Generate

python scripts/generate.py \
  --prompt "..." \
  [--mode gpt-image-1|nano-banana|nano-banana-2] \
  [--size 1024x1024|1536x1024|1024x1536|auto] \
  [--resolution 1K|2K|4K] \
  [--search-mode] \
  [--enhance-mode] \
  [--idempotency-key KEY] \
  [--output PATH]

Edit

Use this when user wants iterative refinement, style transfer, background changes, object replacement, etc.

python scripts/edit.py \
  --prompt "make the sky sunset orange and add volumetric light" \
  (--image-path /path/to/input.png | --image-url https://...) \
  [--mode gpt-image-1|nano-banana|nano-banana-2] \
  [--size 1024x1024|1536x1024|1024x1536|auto] \
  [--resolution 1K|2K|4K] \
  [--search-mode] \
  [--enhance-mode] \
  [--idempotency-key KEY] \
  [--output PATH]

Mode Mapping

gpt-image-1: general quality route
nano-banana: fast generation route
nano-banana-2: advanced generation route

Recommended Workflow

First pass with generate_image_text (scripts/generate.py).
Use edit_image_text (scripts/edit.py) for targeted changes.
Use a new idempotency_key for each new edit intent.

Idempotency

idempotency_key prevents duplicate billing on retries. If omitted, scripts auto-generate one UUID-based key.

References

安全使用建议

Before installing or using this skill: (1) don't set or expose your Filtrix API key globally until you trust the source—the SKILL.md requires FILTRIX_MCP_API_KEY but the registry metadata omits it; verify which env var your deployment will inject. (2) Note the scripts also accept an undocumented MCP_API_KEY fallback—ask the author to document/fix this. (3) Running the scripts will send images (including anything you pass via --image-path or --image-url) to https://mcp.filtrix.ai/mcp and will download returned signed URLs; avoid sending sensitive images unless you trust Filtrix. (4) The package contains plain Python scripts (no install fetches), so inspect them or run them in a sandboxed environment and monitor network/credential usage. (5) Ask the publisher to correct the registry metadata to declare the required env var (FILTRIX_MCP_API_KEY) and clarify the MCP_API_KEY fallback before trusting the skill.

功能分析

Type: OpenClaw Skill Name: filtrix-image-gen Version: 1.0.2 The skill bundle provides a legitimate interface for the Filtrix AI image generation and editing service via a remote Model Context Protocol (MCP) endpoint (mcp.filtrix.ai). The Python scripts (scripts/generate.py and scripts/edit.py) use standard libraries to perform authenticated API requests, handle image data (base64 and URLs), and save results to the local filesystem as described in the documentation. No evidence of malicious intent, data exfiltration, or unauthorized execution was found.

能力评估

ℹ Purpose & Capability

The skill name/description match the included scripts and MCP usage: both generate and edit images via a Filtrix MCP endpoint. However the registry metadata lists no required environment variables or primary credential, while SKILL.md and the scripts clearly require an API key (FILTRIX_MCP_API_KEY). This mismatch is inconsistent and should be corrected.

ℹ Instruction Scope

SKILL.md and the scripts limit behavior to calling the Filtrix MCP endpoint and downloading returned images or user-provided image URLs. The scripts read local image files when provided, base64-encode them, send them to the MCP endpoint, and write result images to /tmp (or a user-specified path). The scripts also accept an undocumented fallback env var (MCP_API_KEY). No instructions attempt to read unrelated system files or other credentials, but the acceptance of arbitrary input image URLs means user-provided URLs (and image contents) will be retrieved and transmitted to the remote MCP service—consider privacy implications.

✓ Install Mechanism

No install spec; code is included as plain Python scripts. This is the lower-risk model (nothing is automatically fetched or executed at install time), but you still run the bundled scripts locally which will make outbound network calls.

⚠ Credentials

The runtime requires a Filtrix API key (FILTRIX_MCP_API_KEY) but the registry metadata omitted this. The scripts also accept MCP_API_KEY as an undocumented alias. Requesting a single service API key is proportionate for this functionality, but the metadata mismatch and undocumented fallback increase the risk of accidental credential exposure or misconfiguration.

✓ Persistence & Privilege

The skill is not always-enabled, does not request unusual system privileges, and does not modify other skills or global agent settings. It runs only when invoked and writes output files to user-specified paths (default /tmp).

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 filtrix-image-gen
安装完成后，直接呼叫该 Skill 的名称或使用 /filtrix-image-gen 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

v1.0.2

- Migrated to Filtrix Remote MCP; now uses a unified endpoint for all image generation and editing. - Added support for gpt-image-1, nano-banana, and nano-banana-2 modes via MCP. - Removed support and documentation for direct OpenAI, Gemini, and fal.ai integration. - Introduced new references for MCP tools and available image generation modes. - Added clearer setup, idempotency key usage, and recommended workflow guidance.

v1.0.1

Initial release

v1.0.0

Initial release

元数据

Slug filtrix-image-gen

版本 1.0.2

许可证 MIT-0

累计安装 3

当前安装数 3

历史版本数 3

常见问题