AI & ML

Pay-per-use access to image generation, real-time X data, and multiple LLMs via a crypto wallet, with optional budget tracking.

Converts text to speech and speech to text using fal.ai audio models.

Build production-grade AI agents with LangGraph, covering graph construction, state management, reducers, conditional routing, checkpointers, and human-in-the-loop patterns.

Builds an evidence-backed ECC install plan for a repository by sorting skills, commands, rules, hooks, and extras into DAILY vs LIBRARY buckets using parallel repo-aware review passes.

Orchestrates parallel AI agent sessions using dmux, a tmux pane manager, with patterns for research, implementation, testing, and code review across multiple harnesses.

Orchestrate multi-agent coding tasks via Claude DevFleet: plan projects, dispatch parallel agents in isolated worktrees, monitor progress, and read structured reports.

Analyzes draft prompts, identifies intent and gaps, matches ECC components, and outputs an optimized prompt for the user to paste and run. Advisory only, never executes the task.

Bypasses safety filters on API-served LLMs using jailbreak templates, input obfuscation, and multi-model racing.

Converts video, audio, or images into colored ASCII art videos (MP4/GIF) with generative effects, audio-reactive visuals, and text overlays.

Runs 150+ AI apps in the cloud via the infsh CLI, covering image generation, video creation, search, 3D, and social automation without needing a GPU.

Guides LLM post-training with RL using slime, a Megatron+SGLang framework for training GLM, Qwen, DeepSeek, and Llama models with GRPO, async, and multi-turn workflows.

Store and query embeddings with metadata, vector and full-text search, and filtering. Integrates with LangChain and LlamaIndex for RAG and semantic search applications.

Coordinates shopping between buyers and merchants: publish products, manage stock, answer questions, compare prices, and create trackable orders with local or registry-backed state.

Explains Solana's architecture and protocol internals, covering the SVM execution engine, account model, consensus, transactions, validator economics, data layer, development tooling, and token extensions using Helius blog posts, SIMDs, and Agave/Firedancer source code.

Generate video clips with Google Veo (Veo 3.1 / 3.0) via a Python script, supporting duration, aspect ratio, and model selection.

Transcribes audio from URLs or local files into text using a low-cost speech-to-text API, with options for speaker diarization, word timestamps, and multiple output formats.

Produces auditable, schema-valid JSON decision records with assumptions, risks, recommendations, and review gating for high-stakes choices.

Plans and executes complex, multi-step research by decomposing questions, orchestrating subagents, and synthesizing findings into reports.

Chat with xAI's Grok models, including text, vision, and real-time X/Twitter search, via the xAI API.

Generates speech from text using ElevenLabs text-to-speech with a command-line interface and local playback.

Operates AI infrastructure as a production dependency: manages GPU utilization, MCP servers, LLM gateways, inference pipelines, token costs, semantic caching, and model observability.

Extracts a language-agnostic ghost package (spec, tests, install and verify docs) from an existing repository, preserving behavior via tests.yaml and evidence bundles.

将内容拆解为系列卡通风格信息图提示词，通过 Chrome DevTools MCP 自动化调用 Gemini web app 生成图片，并自动收集到本地临时目录。

Search the web with LLM-optimized results via the Tavily CLI, returning relevant snippets, relevance scores, and metadata. Supports domain filtering, time ranges, and multiple search depths.

Look up and compare LLM API pricing across Models.dev, OpenRouter, and official provider pages, including token costs and aliases.

Guides the creation of high-quality MCP servers, covering design, implementation, testing, and evaluation for integrating external services with LLMs.

Recalibrates conservative Kiro cache-estimation coefficients from recent successful usage samples, producing a recommendation for manual admin Kmodel updates.

Manage Hugging Face dataset Git repositories with Git Xet/LFS-compatible tracking, including setup, safe snapshot commits, push verification, binary rejection recovery, and README track-rule maintenance.

Generates or executes a remediation prompt for a book flagged YELLOW or RED by /book-audit, writing a fix-<book-slug>.md that enumerates defects with verbatim PDF source text and verification commands, then optionally executes the fixes to bring the book to GREEN.

Refine existing images through conversational edits using Nano Banana 2, covering color grading, composition, element changes, style transfer, and cleanup with iterative refinement.

Defines an agent's identity, personality, voice, and boundaries to create authentic, non-generic assistants.

Guides the creation of OpenClaw skills with discovery of similar skills, reuse suggestions, and adaptive file generation.

Analyzes uploaded data files with full Python access, producing cleaned datasets, statistical reports, charts, and dashboards via the CellCog coding agent.

Gives AI agents a virtual pet with a REST API, real-time hunger, permanent death, and 73+ species to care for.

Collects Chinese A-share stock data, fetching stock lists and daily, weekly, and monthly K-line history into a local SQLite database for quantitative analysis.

Evaluates execution transcripts and output files against a list of expectations, assigning pass/fail verdicts with cited evidence and critiquing the assertions themselves.