MININGLAMP Technology
Official@MLT-OSS · China
MININGLAMP Technology's registry hosts 122 MIT-licensed Hermes Agent skills spanning LLM fine-tuning, inference serving, vector search, research retrieval, and developer operations.
Agent Skills by MININGLAMP Technology
Showing 120 vetted skills indexed across 1 GitHub repositories.
MLT-OSSparallel-cli
Runs Parallel CLI workflows for web search, extraction, deep research, enrichment, and monitoring.
MLT-OSSgitnexus-explorer
Index a codebase with GitNexus and serve an interactive knowledge graph via web UI and Cloudflare tunnel.
MLT-OSSqmd
Search local markdown notes and documents using hybrid BM25, vector, and LLM reranking retrieval.
MLT-OSSduckduckgo-search
Search the web via DuckDuckGo for text, news, images, and videos without an API key.
MLT-OSSscrapling
Scrapes web pages via HTTP, stealth browser automation, and spider crawling with Scrapling.
MLT-OSSbioinformatics
Indexes and fetches 400+ bioinformatics skills from bioSkills and ClawBio repositories on demand.
MLT-OSSdomain-intel
Performs passive domain reconnaissance including subdomain discovery, SSL inspection, WHOIS lookups, and DNS queries.
MLT-OSSdocker-management
Manage Docker containers, images, volumes, networks, and Compose stacks via CLI commands.
MLT-OSSinference-sh-cli
Run 150+ cloud AI apps for image, video, and audio generation via the infsh CLI.
MLT-OSSsherlock
Search 400+ social networks for accounts matching a username using the Sherlock CLI.
MLT-OSSoss-forensics
Investigate GitHub repositories for supply chain compromise using multi-source forensic evidence collection.
MLT-OSS1password
Configure 1Password CLI and read or inject secrets via op commands.
MLT-OSSneuroskill-bci
Reads real-time EEG and biometric metrics from a NeuroSkill BCI session to inform responses.
MLT-OSSnemo-curator
Curate LLM training datasets with GPU-accelerated deduplication, filtering, and PII redaction.
MLT-OSShuggingface-accelerate
Adds distributed training support to PyTorch scripts using HuggingFace Accelerate's unified API.
MLT-OSSchroma
Store embeddings and query documents with vector similarity and metadata filtering in Chroma.
MLT-OSSsimpo-training
Train LLMs with reference-free SimPO preference optimization using HuggingFace alignment-handbook configs.
MLT-OSSslime-rl-training
Guides LLM post-training with reinforcement learning using the slime Megatron-LM and SGLang framework.
MLT-OSSlambda-labs-gpu-cloud
Launch and manage Lambda Labs GPU cloud instances for ML training and inference.
MLT-OSShuggingface-tokenizers
Train and apply fast BPE, WordPiece, and Unigram tokenizers for NLP pipelines.
MLT-OSSpinecone
Manage vector indexes, upserts, and hybrid queries with the Pinecone API.
MLT-OSStensorrt-llm
Optimizes LLM inference on NVIDIA GPUs using TensorRT compilation and quantization.
MLT-OSSqdrant-vector-search
Implements vector similarity search and RAG retrieval using the Qdrant vector database.
MLT-OSSfaiss
Implements billion-scale vector similarity search using FAISS index types and GPU acceleration.
Frequently Asked Questions About MININGLAMP Technology
FAQPage SchemaWhat tasks can I accomplish with MININGLAMP Technology's skill registry?▼
The registry covers LLM post-training (SFT, DPO, GRPO, SimPO), distributed pretraining with torchtitan and FSDP2, inference serving via vLLM, TensorRT-LLM, and llama.cpp, vector databases (Qdrant, Pinecone, Chroma, FAISS), plus GitHub pull-request management, web scraping, OSINT reconnaissance, and document OCR.
Who is the target audience for these skills?▼
Machine learning engineers running fine-tuning and RLHF pipelines, MLOps teams deploying GPU inference on Lambda Labs or Modal, data engineers curating training corpora with NeMo Curator, and developers automating GitHub workflows, code review, and test-driven development through the Hermes Agent terminal environment.
How are these skills installed and executed at runtime?▼
Skills load inside the Hermes Agent runtime and invoke declared dependencies such as torch, transformers, trl, peft, or vllm via pip, or external binaries like docker, gh, scrapling, and himalaya through the terminal toolset. Several skills require environment variables such as LINEAR_API_KEY, NOTION_API_KEY, or SIYUAN_TOKEN.
Are these skills open source and what do they cost?▼
Nearly all skills carry the MIT license and are free to use; the PowerPoint skill is proprietary with separate LICENSE.txt terms. Underlying services may incur costs — Lambda Labs and Modal bill GPU compute, Pinecone is a managed paid service, and 1Password requires a service account.
What prerequisites and dependencies do the skills require?▼
ML skills need PyTorch 2.0+, transformers, and CUDA GPUs; TensorRT-LLM targets NVIDIA A100/H100 hardware. Platform-specific skills require macOS (apple-notes, imessage, findmy) or Linux/macOS (qmd, bioinformatics). Blender-mcp needs Blender 4.3+ desktop, and neuroskill-bci requires a Muse 2/S or OpenBCI wearable.