rt-claw

RT-Claw is an OpenClaw-inspired intelligent assistant for embedded devices. It provides an AI chat engine that calls an LLM over HTTP, supports tool-use via function calling (30+ built-in tools), and includes a framework for skills, memory (RAM + persistent NVS/flash), swarm-style node discovery/remote tool invocation, scheduled tasks, and a UART chat-first shell with /commands. It is designed to run across multiple OSAL targets (FreeRTOS, RT-Thread, Linux) and includes networking components (HTTP client, Ethernet/WiFi).

Evaluated Mar 30, 2026 (0d ago)
Homepage ↗ Repo ↗ Ai Ml ai embedded agent tool-use llm freeRTOS rt-thread linux uart swarm skills memory
⚙ Agent Friendliness
33
/ 100
Can an agent use this?
🔒 Security
37
/ 100
Is it safe for agents?
⚡ Reliability
30
/ 100
Does it work consistently?

Score Breakdown

⚙ Agent Friendliness

MCP Quality
0
Documentation
40
Error Messages
0
Auth Simplicity
70
Rate Limits
0

🔒 Security

TLS Enforcement
30
Auth Strength
35
Scope Granularity
20
Dep. Hygiene
40
Secret Handling
60

README mentions setting an API key via an environment variable, which is a better-than-hardcoding default, but it does not document transport security (TLS requirements), how keys are stored/used, scope restrictions, rate limiting, or how errors/logging avoid leaking secrets. Dependency hygiene and CVE posture cannot be determined from the provided content.

⚡ Reliability

Uptime/SLA
0
Version Stability
50
Breaking Changes
40
Error Recovery
30
AF Security Reliability

Best When

You have embedded/edge hardware (or a validated QEMU target) and want a local tool-using AI assistant framework that can be ported across FreeRTOS/RT-Thread/Linux with OS abstraction.

Avoid When

You need a well-specified external API contract (OpenAPI/SDK) or documented security controls (TLS/auth/rate-limit/error codes) at the integration surface without inspecting source/docs.

Use Cases

  • Running an on-device/edge AI assistant that can control hardware tools (GPIO, LCD, audio, etc.)
  • Building multi-step AI workflows via composable “skills” that orchestrate multiple tools
  • Persisting user/system context using device memory (NVS/flash) for long-term recall
  • Distributed/connected embedded setups using swarm discovery and remote tool invocation across nodes
  • Scheduling automated tasks driven by AI (create/list/remove tasks)
  • Interactive development/debugging using a UART REPL that forwards user input to the AI

Not For

  • A full-featured cloud SaaS API platform with standard REST/SDKs for developers who want to integrate quickly from a PC/server
  • Use cases requiring strict, clearly documented enterprise security/compliance guarantees (data handling, auth policies, auditability) without reviewing the underlying implementation
  • High-assurance reliability requirements where retry/idempotency/error semantics are not documented

Interface

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

Authentication

Methods: Environment variable for LLM API key (RTCLAW_AI_API_KEY)
OAuth: No Scopes: No

The README indicates an API key is set via RTCLAW_AI_API_KEY, but does not describe server-side auth, scope granularity, or rotation/auditing mechanisms for RT-Claw itself.

Pricing

Free tier: No
Requires CC: No

Open-source (MIT). Costs would primarily be your LLM provider usage if you connect to external LLM APIs.

Agent Metadata

Pagination
none
Idempotent
False
Retry Guidance
Not documented

Known Gotchas

  • No MCP/standard agent interface documented; integration likely requires wiring into its HTTP/LLM and UART/tool interfaces from the agent runtime you control.
  • Because this targets embedded OSAL platforms, tool invocation and failure modes may depend on hardware drivers and network conditions; error semantics are not documented in the provided material.

Alternatives

Full Evaluation Report

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

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 2026-03-30.

6447
Packages Evaluated
19956
Need Evaluation
586
Need Re-evaluation
Community Powered