# ZeroClaw Configurable agent runtime for messaging, tools and recurring workflows. Canonical profile: https://ki-agentenkompass.de/en/agents/zeroclaw/ Editor: [Michael Meierhoff](https://ki-agentenkompass.de/en/about/) Sources reviewed: 2026-10-16 Basis: provider documentation. Examples are editorial suggestions, not hands-on product tests. ## Overview ZeroClaw is listed as a separate product profile. - Provider: ZeroClaw Labs - Product family: ZeroClaw - Profile type: Self-hosted personal agent - Environment: Own computer or server - Connections: Tools and channels according to configuration - Setup: Install a release and configure the model and agent with Quickstart. - Pricing model: Open source; model and operating costs are separate. ## Key features ### Combine channels Connect messaging, email and webhooks to one agent. Source: https://github.com/zeroclaw-labs/zeroclaw ### Switch models Configure cloud providers or Ollama. Source: https://github.com/zeroclaw-labs/zeroclaw ### Control approvals Supervised mode requests approval for selected actions. Source: https://github.com/zeroclaw-labs/zeroclaw ### Use the dashboard Manage chat, memory, configuration and schedules. Source: https://github.com/zeroclaw-labs/zeroclaw ### Trigger workflows Events can start multi-step workflows. Source: https://github.com/zeroclaw-labs/zeroclaw ## Example workflows ### Try event-driven work - Your input: A test webhook and an approval-gated workflow. - Possible result: A traceable workflow with a controlled action. - What to check: Check content, permissions and actions taken. ### Try local model access - Your input: A running Ollama service and a test task. - Possible result: A result through the selected model connection. - What to check: Record time spent, corrections and costs. ## Requirements and limits ### Setup Requirement: Configure the model, channels and permissions yourself. Source: https://github.com/zeroclaw-labs/zeroclaw ### German in everyday use Not yet confirmed: End-to-end German interaction and output quality have not been tested here. Source: https://github.com/zeroclaw-labs/zeroclaw ### Reliability Not yet confirmed: Success rates for specific workflows have not been measured here. Source: https://github.com/zeroclaw-labs/zeroclaw ## A good fit if … - you want to run the runtime yourself on your own machine - you want to reach one agent through channels such as Discord, Telegram, email or webhooks - you want to set the model and permissions yourself in a configuration file ## Less suitable if … - you want a hosted solution without installation and maintenance - you need organisation-level administrator approvals and roles; the reviewed pages mention none Editorial assessment based on provider documentation. Not a product test. ## Usable AI models Open choice You can configure your own keys from many providers, OpenRouter or an OpenAI-compatible endpoint, including fallback chains. Locally, ZeroClaw runs with Ollama or a self-hosted inference server, for example. - Model families named: Anthropic (Claude), OpenAI (GPT), Google (Gemini), xAI (Grok), Moonshot (Kimi), Alibaba (Qwen), MiniMax, Z.ai (GLM) · Other models can be connected freely - Model choice: You choose - Your own API key: Yes - Router or custom endpoint: Yes - Local models: Yes Documented · Sources accessed on: 2026-11-05 - [ZeroClaw: Provider Catalog](https://github.com/zeroclaw-labs/zeroclaw/blob/master/docs/book/src/providers/catalog.md) - [ZeroClaw: Providers overview](https://github.com/zeroclaw-labs/zeroclaw/blob/master/docs/book/src/providers/overview.md) - [ZeroClaw: README](https://github.com/zeroclaw-labs/zeroclaw) ## Europe & regional availability Installation instructions and model prerequisites reviewed. The assessment covers the local operating path described; complete country and feature availability of all connected services remains open. Reviewed scope: Local software and documented self-hosting; external services assessed separately Reviewed on: 2026-10-19 ### Self-hosting with own model access With requirements ZeroClaw documents a self-operated runtime with Unix and Windows installation. Setup selects a provider, including Ollama and external APIs with their own access requirements. Region covered: Europe: own computer or server; model services are country-specific Documented reason: Technical assessment based on installation and provider selection. These sources do not establish a legal cause for regional restrictions. Official source: https://github.com/zeroclaw-labs/zeroclaw ## Accounts & permissions Based on provider documentation. Permissions depend on the account, integration and organization policies. Review the permissions actually requested before use. Configuration covered: Self-hosted ZeroClaw runtime with your own model keys, channels and gateway Sources reviewed: 2026-11-05 ### Account used Documented: ZeroClaw runs with your own keys. The gateway requires pairing by code by default. A Jira token is stored encrypted; the Microsoft 365 client secret only when the secrets.encrypt option is enabled. Evidence: https://github.com/zeroclaw-labs/zeroclaw Evidence: https://docs.zeroclaw.com/master/en/reference/config.html ### Admin approval Open in reviewed sources: The reviewed pages describe no approval process by organisation administrators. You set permissions yourself in the configuration. Evidence: https://docs.zeroclaw.com/master/en/reference/config.html ### Data access & team sharing Documented with conditions: Permission profiles are deny-by-default, so anything not granted is refused. Browser and HTTP have domain allowlists that by default permit all public hosts, so you must narrow them yourself. According to the provider everything runs on your machine and you own the data. Evidence: https://docs.zeroclaw.com/master/en/reference/config.html Evidence: https://docs.zeroclaw.com/master/en/introduction.html ## Prices and included usage Pricing checked: 2026-10-16 Open source; model and operating costs are separate. Total cost depends on access, usage and configuration. Pricing details: https://github.com/zeroclaw-labs/zeroclaw ## Operation & oversight Based on public provider documentation. Assessments are editorial inferences from the stated properties. Actual time requirements and reliability have not been measured. Deployment covered: ZeroClaw as a self-hosted runtime on your own machine or server Sources reviewed: 2026-11-05 ### Files and working state Documented: Gateway chat sessions are persisted in SQLite by default, and memory uses SQLite by default with retention settings. Configuration is a TOML file, and backups go to state/backups by default. Evidence: https://docs.zeroclaw.com/master/en/reference/config.html ### Sign-ins Documented: Pairing codes are valid for 10 minutes by default, a dashboard session lasts at most 8 hours, and ACP sessions end after an hour without activity. These are default settings. Evidence: https://docs.zeroclaw.com/master/en/reference/config.html ### Starting further runs Documented: Cron jobs can be declared in the configuration. Event-triggered SOPs can start via MQTT, webhook, cron or peripherals. A heartbeat exists and is off by default. Evidence: https://github.com/zeroclaw-labs/zeroclaw Evidence: https://docs.zeroclaw.com/master/en/reference/config.html ### Human intervention Documented: The default level, supervised, requires approval for medium-risk operations and blocks high-risk ones. YOLO mode skips these gates. SOPs have approval gates. Evidence: https://github.com/zeroclaw-labs/zeroclaw ### Detecting errors Documented with conditions: Runtime traces go to a JSONL file with daily rotation. Observation via logs, Prometheus or OpenTelemetry depends on the configured backend. Alert channels for high escalations are possible but empty by default. Evidence: https://docs.zeroclaw.com/master/en/reference/config.html ### Resuming after failures Documented with conditions: According to the provider SOP runs are resumable, and chat sessions persist in SQLite. The reviewed pages did not explain how a specific run is resumed after a failure. Evidence: https://github.com/zeroclaw-labs/zeroclaw Evidence: https://docs.zeroclaw.com/master/en/reference/config.html ### Editorial assessment You run ZeroClaw yourself, so updates, backups, key management and monitoring are your job. The default approvals help, provided you deliberately avoid YOLO mode. ## Memory and conversation continuity Scope: Current agent memory and separate conversation history Sources accessed: 2026-11-04 Explicit memory writes and configurable autosave are separate from prompts and files. ### Storage and scope Documented: Persistent memory uses a configured backend store per agent; files and conversation logs form separate storage areas. - [Memory payload lifecycle](https://raw.githubusercontent.com/zeroclaw-labs/zeroclaw/master/docs/book/src/architecture/memory-payload-lifecycle.md) - [Filesystem components](https://docs.zeroclaw.com/master/en/agents/filesystem.html) ### Selection and timing Documented: Durable entries arise through memory tools or explicitly configured runtime autosave; prompt content is not automatically stored thereby. - [Memory payload lifecycle](https://raw.githubusercontent.com/zeroclaw-labs/zeroclaw/master/docs/book/src/architecture/memory-payload-lifecycle.md) ### Inspect, edit and delete Partly documented: Tools support saving and retrieval; complete correction and deletion workflows are not explained in these sources. - [Memory payload lifecycle](https://raw.githubusercontent.com/zeroclaw-labs/zeroclaw/master/docs/book/src/architecture/memory-payload-lifecycle.md) ### Agent and channel boundaries Documented: Memory is agent-scoped; additional peer access is explicit. Conversation memories also have session filters depending on access path. - [Memory payload lifecycle](https://raw.githubusercontent.com/zeroclaw-labs/zeroclaw/master/docs/book/src/architecture/memory-payload-lifecycle.md) ### Retrieval Documented: At turn start, relevant entries can appear as bounded memory context; filters differ between access paths. - [Memory payload lifecycle](https://raw.githubusercontent.com/zeroclaw-labs/zeroclaw/master/docs/book/src/architecture/memory-payload-lifecycle.md) ### Session reset and compaction Partly documented: Long-term memory is separate from session history. History trimming changes context lossily but does not delete memory; reset details remain unspecified. - [Memory payload lifecycle](https://raw.githubusercontent.com/zeroclaw-labs/zeroclaw/master/docs/book/src/architecture/memory-payload-lifecycle.md) ## Context Configure the model, channels and permissions yourself. Provider sources reviewed; examples are editorial. No hands-on tests. ## Open points Requirements and limits, not yet confirmed: - German in everyday use - Reliability ## Sources - https://github.com/zeroclaw-labs/zeroclaw ## Provider Publishes ZeroClaw and the linked product information. - Operator: ZeroClaw Labs - Organization: Open-source project - Website: https://github.com/zeroclaw-labs/zeroclaw - Reviewed: 2026-10-16 - Provider source: https://github.com/zeroclaw-labs/zeroclaw [Sources and method](https://ki-agentenkompass.de/en/sources/)