Multimodal document, OCR, chart and report tools for Qizhixi
Wellknown found it in public sources; nobody has proven control of it yet. Claiming takes one click if the repository is under your GitHub account, or a small file on your domain otherwise. Verified owners get the badge, 15-minute checks, status alerts, edits that outrank crawled data, and a ranking boost.
Agents can do it too: POST https://wellknown.network/api/v1/claims with {"agent":"qizhixi-mcp","method":"well_known_file"} — machine-readable steps at claim.json, guide at /docs/claim.
Everything here was measured by our prober or read from a registry. Nothing is self-reported.
Attributed to the source that supplied each field. Treated as claims, not facts.
Qizhixi MCP 0.1.0 Four tools: parse_document, analyze_image, create_chart, export_report. DOCX/PDF extraction and local Chinese OCR preserve source locations and return unverified candidates. Charts consume explicit supplied data; reports contain editable text, images and sources. Existing Excel calculations remain separate. Run: qizhixi-mcp --transport stdio STDIO sends PNG image content and embedded DOCX/JSON resources. Client support must be verified; resource URIs are not web download links. HTTP: set a random QZX_API_TOKEN of at least 32 characters, then run qizhixi-mcp --transport http --host 127.0.0.1 --port 8765 Remote file retrieval is disabled unless QZX_FILE_URL_HOSTS is configured. The full project contains DEPLOYMENT.txt, HOSTED_PLAN.txt, tests and acceptance evidence. Official Xfyun deployment is pending; local tests do not certify cloud installation, platform attachment access or user-visible file downloads. Linux OCR currently requires libGL.so.1 and libglib-2.0.so.0 through OpenCV. Bundled Noto CJK font is redistributed under the included SIL OFL license.
Mapped onto the structured taxonomy from declared text and observed tool names. Confidence shown for derived entries.
Every source is kept verbatim. Field changes are logged as events.