zeroclaw

ZeroClaw is a local-first, single-binary Rust personal AI assistant you run on your own devices. It provides a gateway/control plane with a web dashboard to manage an always-on assistant that can connect to multiple chat/communication channels (e.g., WhatsApp/Telegram/Slack/Discord/Signal/Email/Matrix/IRC/etc.), integrate with hardware peripherals (e.g., ESP32/STM32/Arduino/Raspberry Pi GPIO), and run autonomous/"hands" workflows, SOPs, and tools (including mentions of MCP and many integrations).

Evaluated Mar 29, 2026 (0d ago)
Homepage ↗ Repo ↗ Ai Ml ai-ml agents personal-assistant self-hosted messaging automation tools rust
⚙ Agent Friendliness
38
/ 100
Can an agent use this?
🔒 Security
56
/ 100
Is it safe for agents?
⚡ Reliability
32
/ 100
Does it work consistently?

Score Breakdown

⚙ Agent Friendliness

MCP Quality
30
Documentation
55
Error Messages
0
Auth Simplicity
55
Rate Limits
50

🔒 Security

TLS Enforcement
70
Auth Strength
75
Scope Granularity
40
Dep. Hygiene
45
Secret Handling
45

README highlights several security features: DM pairing/allowlist by default for unknown senders, explicit opt-in for public inbound DMs, autonomy levels with approval gating, sandboxing layers (path traversal blocking, command allowlisting, forbidden paths like /etc,/root,~/.ssh), and rate limiting/cost caps. However, the excerpt does not provide detailed guidance on TLS configuration requirements, token storage/redaction practices, precise scope granularity, or dependency vulnerability management.

⚡ Reliability

Uptime/SLA
10
Version Stability
55
Breaking Changes
35
Error Recovery
30
AF Security Reliability

Best When

You want an on-device, always-on personal assistant that can manage many communication channels and tools under configurable autonomy and sandbox policies.

Avoid When

You need a simple, documented public API surface for programmatic third-party clients, or you cannot control/monitor local security settings for DM pairing/allowlisting and tool access.

Use Cases

  • Personal AI assistant that lives on your own hardware and stays online
  • Multi-channel personal inbox/assistant for DMs and inbound messages
  • Local autonomous workflows/SOPs with approvals and sandbox/policy controls
  • Managing agent actions and memory via a local web dashboard
  • Connecting external devices/peripherals to agent workflows
  • Integrating with common LLM providers via OAuth/API-key auth profiles with model/provider failover

Not For

  • Untrusted multi-user deployments without strong isolation and DM policy controls
  • Use as a hosted SaaS where uptime/SLA is required
  • Scenarios requiring a standardized public REST/GraphQL API contract for third-party developers
  • Environments where you cannot safely operate an always-on process that can access messaging channels and tool execution

Interface

REST API
No
GraphQL
No
gRPC
No
MCP Server
No
SDK
No
Webhooks
No

Authentication

Methods: OAuth (provider-specific, including OpenAI Codex subscription; Gemini via Google OAuth; Anthropic API key or auth token) API keys / auth tokens for providers (Anthropic; and installer example suggests passing an API key)
OAuth: Yes Scopes: No

Auth model includes subscription-based OAuth for some providers and API key/token auth for others; documentation suggests provider/profile rotation and failover, but the README excerpt does not specify fine-grained OAuth scopes or how tokens are scoped/rotated internally.

Pricing

Free tier: No
Requires CC: No

Pricing for ZeroClaw itself is not described in the provided README excerpt; it likely depends on your selected LLM/provider usage and any third-party channel integrations.

Agent Metadata

Pagination
none
Idempotent
False
Retry Guidance
Not documented

Known Gotchas

  • ZeroClaw is an always-on, multi-channel agent; inbound DMs should be treated as untrusted and require pairing/allowlisting per the documented default DM policy.
  • Autonomy levels (ReadOnly/Supervised/Full) and sandbox/policy controls are critical; agents should respect and not bypass these controls.
  • The README excerpt mentions many tools and MCP, but the provided content does not expose a clear, machine-consumable MCP tool schema or public agent SDK contract.

Alternatives

Full Evaluation Report

Comprehensive deep-dive: security analysis, reliability audit, agent experience review, cost modeling, competitive positioning, and improvement roadmap for zeroclaw.

AI-powered analysis · PDF + markdown · Delivered within 30 minutes

$99

Package Brief

Quick verdict, integration guide, cost projections, gotchas with workarounds, and alternatives comparison.

Delivered within 10 minutes

$3

Score Monitoring

Get alerted when this package's AF, security, or reliability scores change significantly. Stay ahead of regressions.

Continuous monitoring

$3/mo

Scores are editorial opinions as of 2026-03-29.

5347
Packages Evaluated
21056
Need Evaluation
586
Need Re-evaluation
Community Powered