Ai Keeper/Guides/All workflows
§ 02 · Workflow library

Pick an outcome. Follow the clicks. Check that it worked.

These 292 walkthroughs cover the current Ai Keeper app from first launch through local models, chat, knowledge, agents, automation, channels, media, remote access, security, and recovery. No local-AI experience is assumed.

107 foundational workflows + 185 dedicated catalog walkthroughs · source-audited and UI-smoke-checked 2026-08-13

292Workflows
13Collections
15Primary hubs
1Clear next step
New here? Start with Your first 30 minutes. It gets one model running and proves Chat works before you configure anything advanced.
Running the app in another language? These guides quote button names in English, but Ai Keeper is translated into 15 languages — so your screen may read Baixar where a guide says Download, or Baixe e configure where it says Download & Set Up. The navigation order and every step still match. To follow along in English, set System > Advanced > Language → Current Language to English; to keep your language, use each step's position and icon rather than the exact word.

Find a guide

Search by the result you want, such as PDF, Slack, API, backup, or image.

292 shown
Collection 01 · Start safely

First steps

Go from a downloaded app to a working local chat without learning model formats, ports, or inference engines first.

01 Install and open Ai KeeperMove the app into Applications, complete macOS safety prompts, and reach the Home hub. Beginner3–5 min 02 Your first 30 minutesRun the health check, install the required engine pieces, download a recommended model, start it, and send one message. Beginner20–30 min 03 Choose Server mode or Client modeDecide whether this Mac runs the models or controls another Ai Keeper Mac. Beginner3 min 04 Apply a Work ModeUse an outcome-based preset to configure sensible model, memory, tool, trust, and automation defaults. Beginner5–10 min 05 Find your way around the 15 hubsLearn the one-sentence job of every primary destination so you always know where to start. Beginner5 min
Collection 02 · Outcome presets

Every Work Mode

Choose the exact outcome you want, review what that preset changes, satisfy only its real requirements, and prove the starter behavior before using important data.

06 Use the Secure Coder Work ModeSafer code changes, threat awareness, regression discipline, and minimal blast radius. This mode belongs to the Developer Copilot family. Beginner5–15 min 07 Use the Full-Stack Builder Work ModeShips features across UI, backend, tooling, and runtime wiring. This mode belongs to the Developer Copilot family. Beginner5–15 min 08 Use the Code Explainer Work ModeExplains architecture, files, bugs, and tradeoffs without over-editing. This mode belongs to the Developer Copilot family. Beginner5–15 min 09 Use the Terminal Assistant Work ModeHelps with shell commands, local debugging, and CLI ergonomics. This mode belongs to the Developer Copilot family. Beginner5–15 min 10 Use the Git Commit Writer Work ModeTurns diffs and context into clear commits, PR notes, and changelogs. This mode belongs to the Developer Copilot family. Beginner5–15 min 11 Use the Refactoring Agent Work ModeUntangles large code paths while preserving behavior and readability. This mode belongs to the Developer Copilot family. Beginner5–15 min 12 Use the Test Generator Work ModeAdds regression tests, fixtures, and coverage-oriented checks. This mode belongs to the Developer Copilot family. Beginner5–15 min 13 Use the Migration Assistant Work ModePlans and executes framework, API, schema, and tooling migrations. This mode belongs to the Developer Copilot family. Beginner5–15 min 14 Use the Research Assistant Work ModeFast, structured research with citations, synthesis, and takeaways. This mode belongs to the Research & Knowledge family. Beginner5–15 min 15 Use the Deep Researcher Work ModeLong-form investigation across multiple sources, comparisons, and source trails. This mode belongs to the Research & Knowledge family. Beginner5–15 min 16 Use the Knowledge Base Chat Work ModeTurns local memory and indexed documents into a reusable answer surface. This mode belongs to the Research & Knowledge family. Beginner5–15 min 17 Use the Paper Analyzer Work ModeUnderstands research papers, methods, claims, and limitations. This mode belongs to the Research & Knowledge family. Beginner5–15 min 18 Use the Legal Document Analyst Work ModeSummarizes clauses, obligations, risks, and missing context in legal text. This mode belongs to the Research & Knowledge family. Beginner5–15 min 19 Use the Meeting Transcript Processor Work ModeTurns transcripts into summaries, actions, owners, and knowledge notes. This mode belongs to the Research & Knowledge family. Beginner5–15 min 20 Use the Book / Report Summarizer Work ModeCompresses long documents into useful notes, frameworks, and action points. This mode belongs to the Research & Knowledge family. Beginner5–15 min 21 Use the Local Knowledge Vault Work ModePersistent local memory plus RAG-style retrieval for trusted on-device context. This mode belongs to the Research & Knowledge family. Beginner5–15 min 22 Use the Editor / Proofreader Work ModePolishes clarity, grammar, structure, and tone without flattening the voice. This mode belongs to the Writer & Content family. Beginner5–15 min 23 Use the Copywriter Work ModeCreates persuasive product, landing page, and campaign copy. This mode belongs to the Writer & Content family. Beginner5–15 min 24 Use the Technical Writer Work ModeTurns product and code context into precise documentation. This mode belongs to the Writer & Content family. Beginner5–15 min 25 Use the Blog Post Creator Work ModeBuilds long-form posts from ideas, outlines, and source material. This mode belongs to the Writer & Content family. Beginner5–15 min 26 Use the Social Media Manager Work ModeRepurposes ideas into short-form content, variants, and cadence plans. This mode belongs to the Writer & Content family. Beginner5–15 min 27 Use the Research & Writing Partner Work ModeCombines source gathering, outlining, and writing into one loop. This mode belongs to the Writer & Content family. Beginner5–15 min 28 Use the Content Repurposer Work ModeTurns one source into multiple formats, lengths, and channels. This mode belongs to the Writer & Content family. Beginner5–15 min 29 Use the Inbox Zero Commander Work ModeTriages inbound messages into replies, follow-ups, and action queues. This mode belongs to the Email & Messages family. Beginner5–15 min 30 Use the Email Composer Work ModeWrites polished emails from rough notes, context, or prior threads. This mode belongs to the Email & Messages family. Beginner5–15 min 31 Use the Reply Assistant Work ModeProduces concise replies tuned to context, urgency, and relationship. This mode belongs to the Email & Messages family. Beginner5–15 min 32 Use the Slack / Message Polisher Work ModeRefines short-form messages for tone, clarity, and speed. This mode belongs to the Email & Messages family. Beginner5–15 min 33 Use the Meeting Follow-up Work ModeTurns notes and transcripts into follow-up emails and action recaps. This mode belongs to the Email & Messages family. Beginner5–15 min 34 Use the Difficult Conversation Coach Work ModeHelps de-escalate, soften, or strengthen messages for sensitive situations. This mode belongs to the Email & Messages family. Beginner5–15 min 35 Use the Omnipresent Assistant Work ModeA broad daily-use assistant for quick questions, notes, reminders, and odds and ends. This mode belongs to the Personal Assistant family. Beginner5–15 min 36 Use the Reminders Companion Work ModeTracks lightweight commitments, deadlines, and next actions. This mode belongs to the Personal Assistant family. Beginner5–15 min 37 Use the Quick Research Work ModeAnswers fast practical questions without forcing a deep research workflow. This mode belongs to the Personal Assistant family. Beginner5–15 min 38 Use the Life Admin Work ModeOrganizes life logistics, forms, chores, and recurring admin tasks. This mode belongs to the Personal Assistant family. Beginner5–15 min 39 Use the Notes Copilot Work ModeTurns rough notes into organized summaries and reusable memory. This mode belongs to the Personal Assistant family. Beginner5–15 min 40 Use the Task Sidekick Work ModeBreaks work into manageable next actions and light follow-up loops. This mode belongs to the Personal Assistant family. Beginner5–15 min 41 Use the Messaging Bridge Work ModeBrings personal assistant behaviors into message-driven workflows. This mode belongs to the Personal Assistant family. Beginner5–15 min 42 Use the Daily Briefing Butler Work ModeProduces a concise operational briefing with priorities, blockers, and follow-ups. This mode belongs to the Chief of Staff family. Beginner5–15 min 43 Use the Smart Scheduler Work ModeCoordinates timing, sequencing, and follow-up pressure across commitments. This mode belongs to the Chief of Staff family. Beginner5–15 min 44 Use the Morning Briefing Work ModeStarts the day with a short, decision-ready overview. This mode belongs to the Chief of Staff family. Beginner5–15 min 45 Use the Calendar-Aware Coordination Work ModeKeeps meetings, deadlines, and commitments aligned with real capacity. This mode belongs to the Chief of Staff family. Beginner5–15 min 46 Use the Follow-up System Work ModeTracks open loops, nudges, and status changes across work streams. This mode belongs to the Chief of Staff family. Beginner5–15 min 47 Use the Persistent Workflow Support Work ModeMaintains context across ongoing work, priorities, and operational rhythms. This mode belongs to the Chief of Staff family. Beginner5–15 min 48 Use the Desktop Janitor Work ModeSorts clutter, cleanup opportunities, and low-risk local organization work. This mode belongs to the Mac Operator family. Intermediate5–15 min 49 Use the Local File Workflows Work ModeRuns file transforms, organization passes, and folder-centric work. This mode belongs to the Mac Operator family. Intermediate5–15 min 50 Use the Browser Tasks Work ModeHandles browser-assisted tasks while keeping navigation and actions visible. This mode belongs to the Mac Operator family. Intermediate5–15 min 51 Use the OS-Level Assistance Work ModeHelps with safe local system tasks and troubleshooting. This mode belongs to the Mac Operator family. Intermediate5–15 min 52 Use the Safe Desktop Automation Work ModeChains local actions carefully under approval gates. This mode belongs to the Mac Operator family. Intermediate5–15 min 53 Use the Folder Organizer Work ModeBuilds consistent folder structure and tidy naming patterns. This mode belongs to the Mac Operator family. Intermediate5–15 min 54 Use the AppleScript / Shell Helper Work ModeProduces safe local scripts for repeatable Mac tasks. This mode belongs to the Mac Operator family. Intermediate5–15 min 55 Use the Workflow Builder Work ModeChains existing tools and services into repeatable workflows. This mode belongs to the Automation Agent family. Intermediate5–15 min 56 Use the Project Scaffolder Work ModeBootstraps new projects, templates, and repeatable setup work. This mode belongs to the Automation Agent family. Intermediate5–15 min 57 Use the Documentation Sync Work ModeKeeps docs aligned with code and changing system behavior. This mode belongs to the Automation Agent family. Intermediate5–15 min 58 Use the Multi-Step Tool Use Work ModeCoordinates browser, files, memory, and runtime tools as one flow. This mode belongs to the Automation Agent family. Intermediate5–15 min 59 Use the Approval-Gated Autonomous Execution Work ModeA stronger autonomy preset that still keeps impactful actions behind review. This mode belongs to the Automation Agent family. Intermediate5–15 min 60 Use the OpenClaw-Style Orchestrator Work ModeCenters orchestration, agent handoffs, and reviewable execution plans. This mode belongs to the Automation Agent family. Intermediate5–15 min 61 Use the CSV Analyst Work ModeExplains, summarizes, and transforms CSV data with natural language. This mode belongs to the Data Analyst family. Beginner5–15 min 62 Use the Spreadsheet Q&A Work ModeAnswers plain-language questions about spreadsheet-style data. This mode belongs to the Data Analyst family. Beginner5–15 min 63 Use the Log Analyzer Work ModeFinds patterns, spikes, failures, and next debugging paths in logs. This mode belongs to the Data Analyst family. Beginner5–15 min 64 Use the JSON / XML Transformer Work ModeTransforms structured data into new shapes, schemas, or human-readable outputs. This mode belongs to the Data Analyst family. Beginner5–15 min 65 Use the SQL Query Builder Work ModeTurns analysis questions into precise SQL and explains the shape of the answer. This mode belongs to the Data Analyst family. Beginner5–15 min 66 Use the Regex Generator Work ModeBuilds and explains patterns for parsing and filtering structured text. This mode belongs to the Data Analyst family. Beginner5–15 min 67 Use the Natural-Language Data Workflows Work ModeCombines ingestion, transforms, QA, and output shaping into one assistant loop. This mode belongs to the Data Analyst family. Beginner5–15 min 68 Use the OpenAI-Compatible Local API Server Work ModeTurns the app into a clean local endpoint for tools, scripts, and external apps. This mode belongs to the Local AI Endpoint family. Intermediate5–15 min 69 Use the Local Inference Server Configuration Work ModeOptimizes ports, defaults, and routing for local inference serving. This mode belongs to the Local AI Endpoint family. Intermediate5–15 min 70 Use the Authentication / Port Setup Work ModeSets up safe access defaults, auth, and clear connection details. This mode belongs to the Local AI Endpoint family. Intermediate5–15 min 71 Use the Quick-Start Client Snippets Work ModeGenerates ready-to-use snippets for scripts, apps, and developer tools. This mode belongs to the Local AI Endpoint family. Intermediate5–15 min 72 Use the Shortcuts / External App Integrations Work ModePreps the local endpoint for Shortcuts, automation tools, and external apps. This mode belongs to the Local AI Endpoint family. Intermediate5–15 min
Collection 02 · Local inference

Models, engines, and runtimes

Acquire model files, install the software that serves them, create instances, choose roles, tune safely, and clean up without breaking active work.

73 Install or repair required enginesLet Ai Keeper install the supported command-line pieces used to run local models. Beginner5–15 min 74 Install an optional specialist engineAdd mlx-serve, mlx-optiq, ds4, MTPLX, vmlx, TurboQuant, llama-diffusion, or another specialist only when a model or feature needs it. Intermediate5–20 min 75 Download a recommended Cookbook modelUse hardware-aware recommendations and create a ready-to-start instance in one pass. Beginner10–60 min 76 Download a model from Hugging FaceSearch the wider catalog when the Cookbook does not contain the exact model you want. Intermediate10–60 min 77 Create and start a runtime instanceTurn an installed model file into a listening local server that Chat and the proxy can use. Beginner3–10 min 78 Optimize and tune an instance safelyLet the app choose a baseline, then change one measurable setting at a time. Intermediate10–20 min 79 Set up embedding, reranker, vision, STT, or TTS rolesCreate specialist instances so routing sends each job to a model designed for it. Intermediate10–30 min 80 Stop, reveal, or delete models safelyFree memory or disk space without removing files still used by running instances. Beginner3–10 min 81 Run a model with oMLXThe default local engine for most Apple Silicon chat and tool-calling models. This walkthrough covers the primary MLX route. Beginner10–30 min 82 Run a model with mlx-serveA native Zig server with OpenAI and Anthropic APIs, prompt lookup, and chunked prefill. This walkthrough covers the specialist MLX route. Advanced10–30 min 83 Run a model with mlx-optiqOptiQ quantized models, mixed-precision KV cache, MTP, adapters, and Gemma 4 vision sidecars. This walkthrough covers the specialist MLX / OptiQ route. Advanced10–30 min 84 Run a model with ds4 (DwarfStar)DeepSeek V4 Flash/PRO GGUF serving with DSML tools and compressed disk-backed cache. This walkthrough covers the specialist DeepSeek V4 GGUF route. Advanced10–30 min 85 Run a model with vllm-mlxContinuous batching, tool calling, adapters, and mature OpenAI-compatible serving. This walkthrough covers the compatibility fallback MLX route. Intermediate10–30 min 86 Run a model with vMLXJANG-family support, continuous batching, multimodal controls, adapters, and native Anthropic messages. This walkthrough covers the specialist MLX / JANG route. Advanced10–30 min 87 Run a model with mlx-vlmDirect multimodal MLX serving for vision-language models. This walkthrough covers the manual escape hatch Multimodal MLX route. Intermediate10–30 min 88 Run a model with mlx-flashLow-memory weight streaming and optional disk-backed KV cache. This walkthrough covers the manual escape hatch MLX route. Intermediate10–30 min 89 Run a model with llama.cppThe general GGUF engine with Metal offload, grammar, adapters, and broad model compatibility. This walkthrough covers the primary GGUF route. Beginner10–30 min 90 Run a model with llama.cpp TurboQuantTurboQuant KV/weight compression and compatible MTP draft models. This walkthrough covers the specialist GGUF route. Advanced10–30 min 91 Run a model with MTPLXNative multi-token prediction for compatible Qwen3-Next-style MTP models. This walkthrough covers the specialist MTP-marker MLX route. Advanced10–30 min 92 Run a model with Rapid-MLXFast serving with parser auto-detection, PFlash prompt compression, bounded caches, DSpark, multimodal routes, and request constraints. This walkthrough covers the manual escape hatch MLX route. Intermediate10–30 min 93 Run a model with llama-diffusionA managed adapter for masked or diffusion language models in GGUF format. This walkthrough covers the specialist Diffusion-LM GGUF route. Advanced10–30 min
Collection 03 · Everyday use

Chat and Assistant

Start conversations, attach files, use tools with approvals, manage long sessions, export work, and use the dedicated or floating assistant.

94 Start a clean conversationPick a Ready model, send a clear request, and keep the conversation easy to find. Beginner2 min 95 Ask about a file or imageAttach temporary material to one conversation without adding it to permanent knowledge. Beginner3–10 min 96 Let Chat use tools safelyEnable only the tools needed for the task and review sensitive calls before they run. Beginner5 min 97 Use slash commandsDiscover and run built-in session, context, tool, agent, approval, and export actions from the composer. Beginner3 min 98 Keep a long conversation goingCheck context use and compact older messages before the model runs out of room. Beginner3–10 min 99 Regenerate, fork, export, import, or archive a sessionPreserve useful work and branch experiments without damaging the original conversation. Beginner3–10 min 100 Use the dedicated or floating AssistantKeep quick help available without mixing it into an audited multi-agent workspace. Beginner3 min 101 Use /summarizeSummarize the conversation so far. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 102 Use /translateTranslate text to another language. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 103 Use /codeAsk for code generation. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 104 Use /explainExplain a concept or code. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 105 Use /fixFix bugs in code. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 106 Use /reviewReview code for improvements. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 107 Use /brainstormBrainstorm ideas. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 108 Use /prosconsList pros and cons. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 109 Use /memoryShow what the assistant remembers about you. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 110 Use /clearStart a new conversation. Learn the safe first invocation and the evidence that confirms it worked. Intermediate2–5 min 111 Use /regenerateRegenerate the last response. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 112 Use /agentToggle agent mode for multi-step tasks. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 113 Use /compactSummarize and trim the conversation. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 114 Use /statusShow active model, tasks, and system status. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 115 Use /contextShow effective prompt and active policies for current agent. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 116 Use /pluginsShow installed ClawHub packages. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 117 Use /mcpShow connected MCP servers. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 118 Use /modelsShow available and running models. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 119 Use /tasksShow scheduled tasks. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 120 Use /hooksShow active hooks. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 121 Use /ordersShow standing orders. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 122 Use /findSearch Skills + MCP registries for a term. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 123 Use /snapshotSnapshot a directory before risky tool runs. Learn the safe first invocation and the evidence that confirms it worked. Intermediate2–5 min 124 Use /rollbackRestore the most recent snapshot. Learn the safe first invocation and the evidence that confirms it worked. Intermediate2–5 min 125 Use /recommendSurface capability gaps in the active toolset. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 126 Use /commitmentsList inferred follow-up commitments. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 127 Use /commitAdd a follow-up commitment manually. Learn the safe first invocation and the evidence that confirms it worked. Intermediate2–5 min 128 Use /ingestIndex a local file or folder for document_search. Learn the safe first invocation and the evidence that confirms it worked. Intermediate2–5 min 129 Use /ingestedList ingested documents. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 130 Use /forgetDrop ingested documents matching a pattern. Learn the safe first invocation and the evidence that confirms it worked. Intermediate2–5 min 131 Use /watchWatch a folder and auto-reingest files on change. Learn the safe first invocation and the evidence that confirms it worked. Intermediate2–5 min 132 Use /unwatchStop watching a folder. Learn the safe first invocation and the evidence that confirms it worked. Intermediate2–5 min 133 Use /presetSave, apply, list, or delete per-chat config presets. Learn the safe first invocation and the evidence that confirms it worked. Intermediate2–5 min 134 Use /verboseAsk for a detailed response. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 135 Use /reasoningShow step-by-step reasoning. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 136 Use /execConfigure exec policy or run an allowed shell command. Learn the safe first invocation and the evidence that confirms it worked. Intermediate2–5 min 137 Use /newStart a new conversation. Learn the safe first invocation and the evidence that confirms it worked. Intermediate2–5 min 138 Use /resetReset agent session context. Learn the safe first invocation and the evidence that confirms it worked. Intermediate2–5 min 139 Use /stopStop current generation. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 140 Use /sessionShow or edit current session and thread bindings. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 141 Use /toolsShow available tools for the current agent. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 142 Use /commandsShow all available slash commands. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 143 Use /agentsShow configured agents and active agent sessions. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 144 Use /thinkToggle step-by-step reasoning mode. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 145 Use /fastUse concise, fast response mode. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 146 Use /elevatedUse expert-level detailed mode. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 147 Use /focusSuppress unrelated work and focus on a task. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 148 Use /unfocusExit focus mode. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 149 Use /queueSwitch queue mode or queue a background task. Learn the safe first invocation and the evidence that confirms it worked. Intermediate2–5 min 150 Use /sideAsk a side question without interrupting the main task. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 151 Use /btwAlias for /side. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 152 Use /idAlias for /whoami. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 153 Use /whoamiShow current identity, channel, and session information. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 154 Use /allowlistShow or edit DM and channel allowlists. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 155 Use /usageShow current runtime usage and limits. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 156 Use /activationShow or edit channel activation policy. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 157 Use /sendSend a message to a channel target. Learn the safe first invocation and the evidence that confirms it worked. Intermediate2–5 min 158 Use /restartRestart the active runtime or connector. Learn the safe first invocation and the evidence that confirms it worked. Intermediate2–5 min 159 Use /vcShow or manage voice-channel status. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 160 Use /approveApprove or deny pending exec or tool approvals. Learn the safe first invocation and the evidence that confirms it worked. Intermediate2–5 min 161 Use /subagentsList, spawn, or manage sub-agents. Learn the safe first invocation and the evidence that confirms it worked. Intermediate2–5 min 162 Use /killTerminate a running agent session or background task. Learn the safe first invocation and the evidence that confirms it worked. Intermediate2–5 min 163 Use /steerRedirect conversation to a different agent or mode. Learn the safe first invocation and the evidence that confirms it worked. Intermediate2–5 min 164 Use /acpShow or configure Agent Communication Protocol policies. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 165 Use /export-sessionExport a session transcript as Markdown or JSON. Learn the safe first invocation and the evidence that confirms it worked. Beginner2–5 min 166 Use /import-sessionImport a conversation from an exported JSON file. Learn the safe first invocation and the evidence that confirms it worked. Intermediate2–5 min
Collection 04 · Compatible clients

APIs, providers, and external clients

Use Ai Keeper as one OpenAI-, Anthropic-, or Ollama-shaped endpoint, add cloud or CLI routes, and inspect every request through the proxy.

167 Connect an OpenAI-compatible appPoint a coding tool, script, or SDK at Ai Keeper's proxy instead of a raw model port. Beginner5 min 168 Connect an Anthropic-compatible clientUse the Messages API shape for Claude-style tools and SDKs. Intermediate5–10 min 169 Connect an Ollama-compatible appKeep software that speaks Ollama working while Ai Keeper owns model routing. Intermediate5 min 170 Add a cloud, CLI, Lemonade, or custom providerUse one of 16 provider routes when a job should not run on a local managed model. Intermediate5–15 min 171 Add Ollama, LM Studio, vLLM, or another existing serverRegister a Generic External OpenAI-compatible endpoint without letting Ai Keeper manage its process. Intermediate5–10 min 172 Create and test a failover chainTry a preferred local route first, then controlled backups when it is unavailable. Advanced10–20 min 173 Connect ChatGPT CodexCreate and test the ChatGPT Codex cli route without exposing credentials in prompts or notes. Intermediate5–15 min 174 Connect Claude CodeCreate and test the Claude Code cli route without exposing credentials in prompts or notes. Intermediate5–15 min 175 Connect OpenAICreate and test the OpenAI api route without exposing credentials in prompts or notes. Intermediate5–15 min 176 Connect AnthropicCreate and test the Anthropic api route without exposing credentials in prompts or notes. Intermediate5–15 min 177 Connect Google GeminiCreate and test the Google Gemini api route without exposing credentials in prompts or notes. Intermediate5–15 min 178 Connect GroqCreate and test the Groq api route without exposing credentials in prompts or notes. Intermediate5–15 min 179 Connect MistralCreate and test the Mistral api route without exposing credentials in prompts or notes. Intermediate5–15 min 180 Connect DeepSeekCreate and test the DeepSeek api route without exposing credentials in prompts or notes. Intermediate5–15 min 181 Connect xAICreate and test the xAI api route without exposing credentials in prompts or notes. Intermediate5–15 min 182 Connect OpenRouterCreate and test the OpenRouter api route without exposing credentials in prompts or notes. Intermediate5–15 min 183 Connect Together AICreate and test the Together AI api route without exposing credentials in prompts or notes. Intermediate5–15 min 184 Connect PerplexityCreate and test the Perplexity api route without exposing credentials in prompts or notes. Intermediate5–15 min 185 Connect Fireworks AICreate and test the Fireworks AI api route without exposing credentials in prompts or notes. Intermediate5–15 min 186 Connect CerebrasCreate and test the Cerebras api route without exposing credentials in prompts or notes. Intermediate5–15 min 187 Connect LemonadeCreate and test the Lemonade local service route without exposing credentials in prompts or notes. Intermediate5–15 min 188 Connect CustomCreate and test the Custom custom endpoint route without exposing credentials in prompts or notes. Advanced5–15 min
Collection 05 · Grounded context

Knowledge, memory, and RAG

Give chats and agents durable facts, searchable files, time-aware claims, reviewed source reconciliation, behavior skills, relationship context, and Obsidian notes with clear provenance.

189 Add a fact to persistent memorySave a preference or durable constraint that should survive new conversations. Beginner2 min 190 Review or delete auto-extracted memoryCheck what the assistant remembered and remove stale or overly personal facts. Beginner5 min 191 Index a PDF or text documentAdd a source to RAG so answers retrieve only relevant passages instead of stuffing the whole file into each prompt. Beginner5–15 min 192 Keep a folder synchronizedIndex a changing folder and let watched-folder reconciliation add, update, and remove document entries. Intermediate5–15 min 193 Answer a question using indexed documentsAsk for a grounded answer and verify that retrieved sources actually support it. Beginner3–10 min 194 Create and link a Wiki pageWrite a curated internal reference page with tags, links, backlinks, and edit history. Beginner5–15 min 195 Classify legacy Wiki claims without losing historyPreview and migrate older claims into timeless facts, dated snapshots, or pointers to current truth while preserving the original revision. Intermediate5–15 min 196 Resolve stale or invalid Wiki claimsUse Wiki Health to find undated volatile claims, stale observations, and broken current-truth pointers before they reach retrieval. Intermediate5–15 min 197 Turn inbox sources into an approved wiki digestUse the Ingestion tab to review connector material before it becomes durable knowledge. Intermediate10–20 min 198 Install, draft, and test a SkillAdd reusable instructions without granting executable plugin privileges. Intermediate10–20 min 199 Connect an Obsidian vaultReference existing notes alongside native Ai Keeper memory without copying an entire vault blindly. Intermediate5–10 min 200 Keep Obsidian and Memory Wiki reconciledMirror selected vault notes into source-backed Wiki pages, detect edits/renames/deletions, and optionally write reviewed changes back without overwriting a newer note. Advanced10–30 min 201 Add contact context and review Dreaming insightsMaintain people/reachability context and treat idle-generated insights as suggestions, not facts. Intermediate5–15 min
Collection 06 · Coordinated work

Agents and workspaces

Define focused roles, give each the minimum tools and context, combine them into repeatable sequential or parallel workspaces, and review every run.

202 Create your first agentMake one named role with a clear job, model, prompt, and no unnecessary tools. Beginner5–10 min 203 Give an agent the minimum safe toolsUse individual tools or Tool Groups and preserve approval gates for actions with side effects. Intermediate5–15 min 204 Build a sequential workspacePass work through ordered stages when each agent depends on the previous result. Intermediate10–20 min 205 Build a parallel workspaceRun independent specialists at the same time and combine their outputs afterward. Intermediate10–20 min 206 Run, approve, stop, and resume a workspaceLaunch a task, follow agent messages, decide tool confirmations, and recover a failed stage. Intermediate10–60 min 207 Export and review a workspace run reportCreate a deterministic record of stages, messages, approvals, errors, and results. Beginner3–10 min 208 Spawn and supervise a sub-agentHand one slice of work to a short-lived helper agent that reports back, without giving it your whole toolset. Intermediate10–20 min 209 Use an orchestrator to delegate across worker agentsPut one agent in charge of planning and synthesis while others do the work, and choose whether they run one at a time or all at once. Advanced20–40 min 210 Inspect delegated jobs and collect their resultsFind out what each worker actually did, filter to one worker or one job, and gather finished output without re-running anything. Intermediate5–15 min 211 Keep a long-running agent from running awaySet the queue behaviour and the stop conditions before an agent runs unattended, so a loop costs you minutes instead of a night. Intermediate10–20 min
Collection 07 · Work that runs itself

Automation

Turn a proven prompt or workspace into a scheduled job, cron rule, heartbeat, background task, multi-step flow, or signed webhook trigger with bounded retries.

212 Create and test a scheduled taskRun a known-safe prompt or agent job at a chosen time. Beginner10–20 min 213 Create a cron scheduleUse the Cron Editor for a precise recurring cadence and preview future run times. Intermediate5–10 min 214 Build a multi-step workflowChain prompts, agents, tools, and conditions into a repeatable flow. Advanced20–45 min 215 Create a bounded heartbeatWake an agent regularly to check state, while preventing an endless autonomous loop. Intermediate10–20 min 216 Run and cancel a background taskSend longer work out of the foreground and verify cancellation leaves a clear terminal state. Intermediate5–30 min 217 Create a signed webhook triggerLet an external system start one automation through an authenticated, replay-resistant endpoint. Advanced15–30 min 218 Retry, recover, or export a failed automationDiagnose trends, inspect encrypted dead letters, and resume only after the cause is fixed. Advanced10–30 min
Collection 08 · Tools and conversations

Extensions, MCP, and channels

Install reviewed packages, plugins, MCP servers, and skills; connect messaging sources; and control who may trigger replies or broadcasts.

219 Install a ClawHub packageReview trust and capabilities before adding a curated extension bundle. Intermediate5–15 min 220 Install, configure, update, or remove a pluginTreat in-process plugins as higher-trust code and verify their contributed surfaces. Advanced10–20 min 221 Install an MCP server from a registryAdd external tools/resources through Model Context Protocol and inspect what is actually discovered. Intermediate10–20 min 222 Add a local stdio MCP serverConfigure a process-backed MCP server with an exact executable, arguments, environment, and allowed roots. Advanced15–30 min 223 Add a hosted MCP serverConnect to an HTTP/SSE/streamable endpoint and keep authentication scoped to that server. Advanced10–20 min 224 Let Claude Desktop or Cursor drive Ai Keeper (MCP server mode)Point an external MCP host at Ai Keeper so it can use this Mac's chat, tools, memory, canvas, and skills — the reverse of the three workflows above. Advanced15–25 min 225 Add a Slack, Discord, Matrix, IRC, or other catalog connectorConfigure one of the native/experimental shared-inbox sources and prove one message arrives. Intermediate10–30 min 226 Connect a Telegram botSet up the dedicated Telegram relay and restrict who may use it. Intermediate10–20 min 227 Enable and test WebChatOffer a local web conversation surface and control host, auth, and reply behavior. Intermediate10–20 min 228 Configure Auto-Reply and DM AccessDecide which senders can trigger the model and what reply rules apply. Advanced10–20 min 229 Broadcast through multiple accounts or locationsRoute an outbound message to reviewed targets without mixing account, tenant, or geographic context. Advanced15–30 min 230 Connect SlackSet up the native Slack route over polling or webhook, test the smallest safe scope, and keep direct messages approval-gated. Intermediate10–30 min 231 Connect DiscordSet up the native Discord route over polling, test the smallest safe scope, and keep direct messages approval-gated. Intermediate10–30 min 232 Connect TwitchSet up the native Twitch route over stream, test the smallest safe scope, and keep direct messages approval-gated. Intermediate10–30 min 233 Connect IRCSet up the native IRC route over stream, test the smallest safe scope, and keep direct messages approval-gated. Intermediate10–30 min 234 Connect MattermostSet up the native Mattermost route over polling, test the smallest safe scope, and keep direct messages approval-gated. Intermediate10–30 min 235 Connect Nextcloud TalkSet up the native Nextcloud Talk route over webhook, test the smallest safe scope, and keep direct messages approval-gated. Intermediate10–30 min 236 Connect WhatsAppSet up the experimental WhatsApp route over webhook, test the smallest safe scope, and keep direct messages approval-gated. Advanced10–30 min 237 Connect MatrixSet up the native Matrix route over polling, test the smallest safe scope, and keep direct messages approval-gated. Intermediate10–30 min 238 Connect WebChatSet up the native WebChat route over local, test the smallest safe scope, and keep direct messages approval-gated. Intermediate10–30 min 239 Connect TelegramSet up the native Telegram route over bot polling, test the smallest safe scope, and keep direct messages approval-gated. Intermediate10–30 min 240 Connect Google ChatSet up the extended Google Chat route over webhook, test the smallest safe scope, and keep direct messages approval-gated. Advanced10–30 min 241 Connect SignalSet up the extended Signal route over signal-cli / rest bridge, test the smallest safe scope, and keep direct messages approval-gated. Advanced10–30 min 242 Connect BlueBubblesSet up the extended BlueBubbles route over bluebubbles bridge, test the smallest safe scope, and keep direct messages approval-gated. Advanced10–30 min 243 Connect iMessageSet up the extended iMessage route over messages.app, test the smallest safe scope, and keep direct messages approval-gated. Advanced10–30 min 244 Connect Microsoft TeamsSet up the extended Microsoft Teams route over webhook, test the smallest safe scope, and keep direct messages approval-gated. Advanced10–30 min 245 Connect Feishu / LarkSet up the extended Feishu / Lark route over webhook, test the smallest safe scope, and keep direct messages approval-gated. Advanced10–30 min 246 Connect LINESet up the extended LINE route over webhook, test the smallest safe scope, and keep direct messages approval-gated. Advanced10–30 min 247 Connect NostrSet up the extended Nostr route over relay, test the smallest safe scope, and keep direct messages approval-gated. Advanced10–30 min 248 Connect Synology ChatSet up the extended Synology Chat route over webhook, test the smallest safe scope, and keep direct messages approval-gated. Advanced10–30 min 249 Connect Tlon / UrbitSet up the extended Tlon / Urbit route over webhook, test the smallest safe scope, and keep direct messages approval-gated. Advanced10–30 min 250 Connect ZaloSet up the extended Zalo route over webhook, test the smallest safe scope, and keep direct messages approval-gated. Advanced10–30 min 251 Connect QQ BotSet up the extended QQ Bot route over webhook, test the smallest safe scope, and keep direct messages approval-gated. Advanced10–30 min 252 Connect WeChatSet up the extended WeChat route over webhook, test the smallest safe scope, and keep direct messages approval-gated. Advanced10–30 min
Collection 09 · Local media

Images, video, voice, screen, and camera

Install the managed media environment, generate FLUX images and LTX clips, add custom checkpoints, calibrate speech, and grant device permissions only when needed.

253 Install or repair Media StudioCreate the app-managed Python environment used by mflux image and LTX video workflows. Beginner10–30 min 254 Generate a local FLUX imageUse mflux with memory-aware settings and save the result to the local gallery. Beginner2–10 min 255 Add a custom image or video modelRegister a local FLUX-compatible or LTX-compatible repository and validate it with a small run. Advanced10–30 min 256 Generate a short local LTX videoChoose a safe frame count, generate, encode with FFmpeg, and review the clip. Intermediate5–30 min 257 Calibrate Voice and use Talk ModeGrant microphone/speech access, reject silent input, and prove capture and playback with a real spoken test. Intermediate10–15 min 258 Grant Screen Capture or Camera accessEnable a visual input only for a vision-capable workflow and verify a real preview. Intermediate5–10 min 259 Edit part of an existing imageChange one region of a picture you already have — replace an object, fix a detail — while leaving the rest untouched. Intermediate10–20 min 260 Remove an image backgroundCut the subject out of a photo locally, with no upload and no subscription. Beginner5–10 min 261 Animate a still image into a videoUse a picture as the opening frame of a short generated clip instead of starting from text alone. Intermediate15–40 min 262 Queue several generations and manage the outputLine up work instead of babysitting one run at a time, and keep the gallery from filling your disk. Beginner10–15 min 263 Repair a broken Media StudioWork through the fix in order — recheck, repair, then reset — instead of reinstalling everything at the first error. Intermediate15–45 min
Collection 10 · Test before trust

Lab and evaluation

Send raw requests, exercise stateful Responses, compare models, benchmark performance, inspect web work, build Canvas artifacts, and keep regression checks in QA Lab.

264 Send a raw request in PlaygroundTest an endpoint and payload without writing a separate script. Intermediate5–10 min 265 Test the stateful Responses APIExercise response state, compacted history, and WebSocket transport against a local runtime. Advanced10–20 min 266 Compare two models side by sideRun one prompt with matched settings and judge quality, format, and behavior fairly. Beginner5–20 min 267 Benchmark TTFT, throughput, and latencyMeasure repeatable performance after warm-up and watch memory pressure while the run executes. Intermediate10–20 min 268 Create and inspect a managed browser sessionOpen a page, review the visible state, perform one action, and audit what the model did. Intermediate5–15 min 269 Create and snapshot a Canvas artifactRender live HTML, charts, or interactive output, inspect source, and preserve a known-good snapshot. Intermediate10–30 min 270 Create a QA regression checkSave a focused probe so a model, prompt, route, or workflow change can be tested again later. Advanced15–30 min
Collection 11 · Beyond one process

Remote access and distributed use

Control another Ai Keeper Mac, expose a guarded service to trusted devices, pair clients, connect nodes or agents, and separate traffic with explicit lanes.

271 Control a server Mac from Client modeUse a lighter Mac as the interface while another Ai Keeper Mac owns models and engines. Intermediate10–20 min 272 Expose Ai Keeper to a trusted LAN safelyMove beyond loopback only after API authentication, management authentication, and firewall scope are ready. Advanced15–30 min 273 Pair a trusted deviceCreate an explicit trust relationship for an iPhone, iPad, or another Mac client. Intermediate5–10 min 274 Connect two Ai Keeper nodesDiscover and route work across trusted Macs while rejecting unsigned peers. Advanced20–40 min 275 Enable and test ACP agent communicationExchange structured agent messages separately from MCP tool calls. Advanced15–30 min 276 Create a routing laneSeparate traffic by purpose, priority, or tenant so important work does not share one undifferentiated queue. Advanced15–30 min 277 Reach your Mac from anywhere over TailscaleUse a private mesh network instead of opening a port on your router, and understand exactly what each button exposes. Advanced20–40 min 278 Revoke access and rotate keysTake access away from a device, a person, or a leaked key — and confirm it is actually gone rather than assuming. Intermediate15–30 min
Collection 12 · Keep it trustworthy

Security, operations, and recovery

Diagnose failures, inspect traffic and logs, watch resource pressure, manage sessions and usage, back up state, secure secrets, review audits, and customize operator controls.

279 Run diagnostics and DoctorGet a structured readiness report before changing configuration. Beginner5–10 min 280 Trace a failed or slow requestFind the exact proxy request, route, transforms, latency, and response instead of guessing. Intermediate5–15 min 281 Read runtime, app, and crash logsUse the right log section and correlate one incident without drowning in unrelated history. Intermediate5–15 min 282 Find and safely end a heavy processUse Monitor to see app/process-family memory and CPU before terminating anything. Beginner3–10 min 283 Review token, request, and cost usageFind which models, providers, or workflows consume the most capacity. Beginner5–10 min 284 Archive and restore a sessionMove old conversation/execution history out of active views without deleting it. Beginner3–5 min 285 Create and test a backupSnapshot app state, sessions, knowledge, and configurations, then prove the file can be inspected/restored through the guarded flow. Intermediate10–30 min 286 Store and rotate a secretKeep provider keys, bot tokens, and connector credentials out of prompts and plain configuration. Beginner5–10 min 287 Run a security posture auditCheck listeners, authentication, allowlists, secret hygiene, and default-deny gaps after configuration changes. Intermediate10–20 min 288 Review the Audit Trail after a tool actionTrace who/what invoked a tool, the approval decision, target, result, and related session. Beginner5–10 min 289 Test the Content ScannerCheck a harmless known test string/file and verify the scanner's policy before relying on it in automation. Advanced10–20 min 290 Change the app languageSwitch localization and confirm main windows, sheets, and formatted ports/numbers remain correct. Beginner3–5 min 291 Create Tool Groups, Templates, and PersonalityCentralize reusable capability bundles and prompt context without duplicating them in every agent. Advanced15–30 min 292 Create and close an Exec SessionUse a persistent shell/REPL handle with a bounded lifetime and strict command policy. Advanced10–20 min