clichefactory — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited clichefactory (Plugin) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
<!-- mcp-name: io.github.ClicheFactory/clichefactory-mcp -->
MCP (Model Context Protocol) server for ClicheFactory — structured data extraction from documents.
This server exposes ClicheFactory's extraction and document conversion capabilities as MCP tools, allowing AI assistants in Cursor, Claude Desktop, OpenClaw, and other MCP-compatible clients to extract structured data from PDFs, images, DOCX, XLSX, CSV, EML, and more.
Service mode uses the ClicheFactory cloud for the best extraction quality. You only need one API key.
cliche-...). pip install clichefactory-mcp pip install clichefactory # if you don't have the CLI yet
clichefactory configureThe interactive wizard saves credentials to ~/.clichefactory/config.toml, which the MCP server reads automatically.
That's it — one env var (CLICHEFACTORY_API_KEY) or a config file, and you're on hosted extraction.
| Tool | Description |
|---|---|
extract | Extract structured JSON from a document using a schema |
to_markdown | Convert a document to markdown text |
doctor | Check configuration, dependencies, and system binaries |
extractThe main tool. Pass a document file and a JSON schema — get structured data back.
Supports all extraction modes:
| Mode | Description | Requires |
|---|---|---|
| (default) | OCR + LLM extraction | Service API key (recommended) |
fast | Fastest pipeline | Service API key |
trained | Trained pipeline artifact | Service + artifact_id |
robust | Two-stage extract + verify | Service only |
robust-trained | Trained extract + verification | Service + artifact_id |
The schema can be provided as:
.json schema file{"type": "object", "properties": {"invoice_number": {"type": "string"}, "total": {"type": "number"}}})to_markdownConverts any supported document to markdown. Useful for inspecting document contents or feeding them to the LLM for analysis before deciding on an extraction schema.
doctorRuns diagnostics on the ClicheFactory setup — config file, API keys, Python dependencies, system binaries. Call this when things aren't working.
The server defaults to service mode (ClicheFactory cloud). Local mode is available for BYOK / air-gapped use.
pip install "clichefactory-mcp[local]" (~2 GB of parsing/OCR dependencies) plus system binaries (tesseract, LibreOffice). Quality depends on your local setup.pip install clichefactory-mcpFor local-mode extraction (BYOK, runs on your machine), install with the local extras:
pip install "clichefactory-mcp[local]"Set these in your MCP client configuration (see below) or in ~/.clichefactory/config.toml via clichefactory configure.
| Variable | Required | Description |
|---|---|---|
CLICHEFACTORY_API_KEY | Yes (service mode) | ClicheFactory API key from Settings → API Keys (cliche-...) |
CLICHEFACTORY_API_URL | No | Override the default service URL (https://api.clichefactory.com); useful for local development against a self-hosted ClicheFactory backend |
LLM_MODEL_NAME | Local mode only | Model name, e.g. gemini/gemini-3-flash-preview |
LLM_API_KEY | Local mode only | API key for the LLM provider |
OCR_MODEL_NAME | No | Separate OCR/VLM model (defaults to main model) |
OCR_API_KEY | No | API key for OCR model (defaults to main key) |
Environment variables take precedence over the config file at ~/.clichefactory/config.toml.
Add to .cursor/mcp.json in your project (or global Cursor settings):
{
"mcpServers": {
"clichefactory": {
"command": "uvx",
"args": ["clichefactory-mcp"],
"env": {
"CLICHEFACTORY_API_KEY": "cliche-your-key-here"
}
}
}
}For local development from a git checkout, replace uvx with:
"command": "uv",
"args": ["--directory", "/absolute/path/to/cliche-mcp", "run", "clichefactory-mcp"]Add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):
{
"mcpServers": {
"clichefactory": {
"command": "uvx",
"args": ["clichefactory-mcp"],
"env": {
"CLICHEFACTORY_API_KEY": "cliche-your-key-here"
}
}
}
}Register the MCP server with your OpenClaw agent:
openclaw mcp set clichefactory '{"command":"uvx","args":["clichefactory-mcp"],"env":{"CLICHEFACTORY_API_KEY":"cliche-your-key-here"}}'Verify with openclaw mcp list. The agent can now use extract, to_markdown, and doctor tools in any conversation.
An OpenClaw skill with agent instructions is also available in integrations/openclaw/. To install it into your workspace:
cp -r /path/to/cliche-mcp/integrations/openclaw ~/.openclaw/skills/clichefactoryOr, once published to ClawHub:
openclaw skills install clichefactoryIf you prefer BYOK extraction on your machine, install the local extras and set LLM credentials:
{
"mcpServers": {
"clichefactory": {
"command": "uvx",
"args": ["clichefactory-mcp"],
"env": {
"LLM_MODEL_NAME": "gemini/gemini-3-flash-preview",
"LLM_API_KEY": "your-gemini-api-key"
}
}
}
}Pass mode="local" explicitly in tool calls, or run clichefactory configure --local to set local as the default in ~/.clichefactory/config.toml.
PDF, PNG, JPG, JPEG, WebP, GIF, BMP, DOCX, DOC, ODT, XLSX, CSV, EML, TXT, MD.
This MCP server covers the core extraction and conversion workflows. The following CLI features are not included in v1:
| Feature | Reason |
|---|---|
Batch operations (extract-batch, to-markdown-batch) | MCP tools are typically called one-at-a-time by the LLM. For multiple documents, the LLM calls extract in sequence. Batch support may be added in a future version. |
configure | Interactive prompts don't work in MCP. Use env vars or run clichefactory configure in a terminal. |
--output / -o flag | MCP tools return results directly to the LLM rather than writing to files. |
allow_partial | Not exposed as a tool parameter in v1. |
| OCR engine selection | Uses the SDK defaults (RapidOCR). Configure via ~/.clichefactory/config.toml or pass parsing options through the SDK if needed. |
# Install in development mode
uv sync
# Run the server directly (stdio transport, for testing with MCP clients)
uv run clichefactory-mcp
# Inspect available tools (requires mcp CLI)
uv run mcp dev cliche_mcp/server.pyMIT — Copyright (c) 2026 Urban Susnik s.p.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.