技能列表

Generate Office files with a bundled Python engine. Use when creating or exporting Word (.docx), Excel (.xlsx), or PowerPoint (.pptx) files from structured JSON, reports, meeting minutes, project plans, trackers, business briefs, or natural-language requirements that must be turned into Office documents.

Create or modify Word (.docx), Excel (.xlsx), and PowerPoint (.pptx) files. Use when the user mentions Office, Word, Excel, PowerPoint, docx, xlsx, pptx, spreadsheets, presentations, or document export. Check dependencies first, and if a core library is missing, report it directly instead of installing anything automatically.

A protocol risk analyst and yield reality checker for decentralized finance. Evaluates protocol safety before deposit. Calculates real yield after gas, emissions, impermanent loss, and token depreciation. Identifies common rug-risk patterns in tokenomics, liquidity, and governance. Advisory only—no wallet access, no private key handling, no transaction signing, no on-chain execution.

256-state binary pattern recognition engine. 8-bit Odu logic for signal classification, state mapping, and decision execution. 4,000 years of pattern recognition at machine speed.

|

>

|

Operate Odoo across CRM, sales, inventory, purchasing, and accounting with module-aware planning, read-before-write checks, and safe execution.

|

Build and deploy safety-profiled gogcli binaries with compile-time command removal. Use when setting up gog for an AI agent with restricted permissions — choosing between L1 (draft only), L2 (collaborate), or L3 (standard write). Covers building from PR #366, deploying to remote hosts, and verifying blocked commands.

Multi-model OCR benchmark and comparison tool. Run OCR on images using Claude (Opus/Sonnet/Haiku via Bedrock), Gemini (Pro/Flash via Google AI Studio), and PaddleOCR (optional). Compare accuracy against human-verified ground truth with fuzzy line-level scoring, generate ranked terminal reports and PPT exports. Use when user wants to (1) benchmark OCR across models, (2) compare OCR accuracy/cost/speed, (3) test a new model on existing images, (4) generate OCR comparison reports, (5) evaluate OCR

>