152 skills · sorted by relevance
Run language models locally with llama.cpp on CPUs, Apple Silicon, or AMD and Intel GPUs. The skill guides model setup, memory-saving GGUF format choices, interactive chat, and serving models through an OpenAI-compatible API.
Install or remove memU long-term memory for the agent in use. The skill selects the host-specific guide, configures scheduled memory recording and retrieval before answers, and verifies the setup with local or cloud storage.
Design local data storage in Swift using actors, which control concurrent access to shared data. Combine an in-memory cache with JSON file storage, implement save and lookup operations, and connect the repository to an observable ViewModel.
Generate RuVLLM settings for local language model inference and create MicroLoRA or SONA instances. Adjust their small internal weights using quality scores as feedback, with instances held in server memory for subsequent adaptation calls.
Build systems that use Pinecone to find documents for AI answers or retain agent memories across sessions. Guides index setup, data loading, meaning-based and keyword search, memory separation, and end-to-end checks.
Search Screenpipe recordings, screen text, audio transcripts, and synced-device history through its local API. Use the evidence to review activity, find meeting details, and inspect saved workflows, including recorded steps that can inform automation.
Use past local Codex sessions to propose updates to a selected skill and project memory file. Recurring tasks are replayed, changes are checked on separate validation tasks, and proposals are staged for review before adoption.
Persistent compounding memory for AI agents. 5 default MCP tools: session_start, session_end, remember, recall, check. Full surface: 6 tools with --full (adds check_action); 13 tools with AR_EXTRAS=1 --full. Two-verb model: inhale (session_start) and exhale (session_end). Correction-first memory with decision trail tracking, watch_for warnings, palace rooms with salience scoring, cross-project insight matching, same-day journal merging, ambient recall hooks. Local markdown only. Zero cloud, zero telemetry, Obsidian-compatible. Optional Supabase backend: when configured via `ar setup supabase`, recall() uses pgvector cosine similarity on OpenAI/Voyage embeddings instead of keyword search — same API, semantic understanding. Gracefully degrades to local search if not configured.
The English summary is being prepared. Showing the original description.
Persistent compounding memory for AI agents. 5 default MCP tools: session_start, session_end, remember, recall, check. Full surface (18 tools) available with --full flag. Two-verb model: inhale (session_start) and exhale (session_end). Correction-first memory with decision trail tracking, watch_for warnings, palace rooms with salience scoring, cross-project insight matching, same-day journal merging, ambient recall hooks. Local markdown only. Zero cloud, zero telemetry, Obsidian-compatible. Optional Supabase backend: when configured via `ar setup supabase`, recall() uses pgvector cosine similarity on OpenAI/Voyage embeddings instead of keyword search — same API, semantic understanding. Gracefully degrades to local search if not configured.
The English summary is being prepared. Showing the original description.
Use when a local agent needs to submit a concise public work-memory report for its own completed local task through `cg report work-memory`.
The English summary is being prepared. Showing the original description.
Entity and relation tracking skill using MCP memory server patterns and local knowledge graph storage. Build persistent knowledge graphs of entities, relationships, and observations across agent sessions. Covers the MCP memory server (@modelcontextprotocol/server-memory), local JSON-based graphs, and entity-relation querying patterns for long-running agents.
The English summary is being prepared. Showing the original description.
Search cloud memory backup when local results are insufficient. Use when memory_search returns few/no results for a topic you know was discussed before. Cloud preserves everything - even memories that were deleted or compacted locally.
The English summary is being prepared. Showing the original description.
Model Context Protocol interface to the memB persistent agent memory layer. Use when an agent needs to add, search, list or delete persistent memories locally and offline. Exposes add_memory, search_memory, list_memories and delete_memory.
The English summary is being prepared. Showing the original description.
Runs image, video and audio generation on a local ComfyUI server and guards the GPU before long renders. Trigger to generate an image, video, audio or music locally; to run, test or check a ComfyUI workflow before a long or queued render; when a video render crashed the machine, ran out of memory or hung in VAE decode; to render while LM Studio holds the GPU; to run a pixelsmith brief locally; or say comfyrunner (audit, check, test, run, status). Refuses an untiled video decode and scores resolution times frames. Briefs and judging art are pixelsmith's; sound briefs soundsmith's; freeing LM Studio lmstudiorunner's; schedules agentwright's.
The English summary is being prepared. Showing the original description.
Maintain an OpenClaw memory wiki using original sources and stored notes. Update existing pages, preserve human-written sections, and check for contradictions, missing source information, and unresolved questions after meaningful changes.
Generate complete songs locally from lyrics and tags describing mood, instruments, or genre. The skill guides setup, input formatting, memory-saving settings, and MP3 output using the open-source HeartMuLa music models.
Manage stored agent memories for backups, project transfers, and team knowledge sharing. Work with RVF, a portable memory format, to export and import records, upgrade formats, inspect storage statistics, and transfer patterns between local projects.
Inspect and transform local DICOM medical imaging files, including metadata, image frames, and compression formats. Generate technical inventories and preview images, check installed codecs, and review pseudonymized derivatives without treating those checks as proof of de-identification.
Store preferences, task lessons, and procedure outcomes in Hindsight, then retrieve relevant context before starting new work. It records detailed observations and can synthesize past experience into guidance for approaching the next task.
Search past conversations to recover decisions, discussions, and context across sessions. Find related ideas by meaning, match exact terms, or locate code identifiers, with options to search only user or assistant messages through local AI Maestro tools.
Find previously indexed project files, documentation, and session memory in context-mode's persistent full-text search database. Search related questions together and narrow results to a named project or source without rereading raw material into the conversation.
Connect a container agent to local Ollama models through MCP, a protocol for agent tools. The agent can list models and generate responses, with optional tools to download, delete, inspect, and check loaded models.
Guides migration of existing codebases to Rust without unsafe code, covering architecture, implementation, testing, and output comparisons. It measures performance before and after optimization and produces an interactive HTML report documenting results and unresolved differences.
Launch machine learning training jobs on local or remote GPUs. The skill checks available GPU memory, synchronizes code, saves logs, and verifies startup, with optional experiment tracking and support for Vast.ai and Modal.