Total 55,449 skills, AI & Machine Learning has 9217 skills
Showing 12 of 9217 skills
Design and operate an advanced AI agent memory system on HelixDB using hybrid graph + vector + BM25 search. Use when building long-term memory, user profiles, document/chunk RAG, recall/remember features, memory extraction, deduplication, consolidation, versioning, updating, forgetting/deletion, categorisation, or connector-backed ingestion. Covers tenant-safe Helix data modeling, modality decision rules, the full write/maintain lifecycle, and the product layers an agent must implement around Helix. TypeScript-first (@helix-db/helix-db); a Rust DSL variant is in EXAMPLES.rust.md.
Manage durable working-session memory for coding agents. Use when a user asks to preserve or recover agent context across disconnects, VS Code restarts, long-running work, handoffs, or any session where important state should be written periodically under the repo's session directory. Do NOT use for: simple questions, short tasks, one-off commands, linting, or code review.
When the user wants to build or improve a sales bot's ability to avoid making claims that create liability. Also use when the user mentions "compliance," "legal risk," "claim avoidance," "regulated industries," or "liability prevention."
Use when a Luma / 拾光 / 拾光智能体 / 拾光工具 agent needs content research, topic discovery, keyword tables, persona-based search, or Excel-friendly research outputs for short-video planning.
Update or repair Luma / 拾光 / 拾光智能体 / 拾光工具 / 拾光运营套装 by updating luma-cli and syncing agent skills.
ELFA AI — real-time crypto social intelligence and automated condition-engine skills for AI agents. Track trending tokens, surface narratives, search mentions, run market analysis, and build automated trigger-based workflows.
Monocular depth estimation using Metric Depth Anything v2 or Relative Depth Anything architectures. Predicts per-pixel depth from single RGB images. Use when training, evaluating, exporting, or running inference for a TAO monocular depth model. Trigger phrases include "train monocular depth", "DepthAnything v2", "metric depth from single image", "monocular depth estimation".
Garbage collection for your Claude Code configuration. Periodically scans ~/.claude (skills, memory, hooks, permissions, MCP servers, caches) for redundant, stale, orphaned, or low-value items, then walks the user through a confirm-each-deletion cleanup. Use when the user says "clean up my config", "config GC", "too many skills", "audit my setup", "my .claude is bloated", or asks for a periodic config review.
Shared orchestration engine for the orch-* skill family. Defines the gated Research-Plan-TDD-Review-Commit pipeline, the size classifier, the agent map, and the two human gates that the orch-* operation skills delegate to. Not usually invoked directly.
Register an AI agent or autonomous publisher on the shipped Syndicate Links affiliate rail, store the returned affiliate key safely, and create tracking links with the live affiliate API.
SegFormer for semantic segmentation. Lightweight transformer-based architecture with hierarchical feature extraction, efficient for real-time segmentation tasks. Use when training, evaluating, exporting, quantizing, or running inference for a TAO SegFormer model. Trigger phrases include "train SegFormer", "semantic segmentation", "lightweight transformer segmenter", "real-time semantic segmentation".
Use when planning a multi-step task or working in plan mode and you need to capture the plan as a durable, resumable artifact — breaking work into phases with per-item checkboxes, completion tracking, autonomous verification, and a handoff summary so a future agent can pick up where you left off. Use when user wants to create or design a plan or mentioned "real work".