agentrove

Agentrove is a self-hosted Claude Code workspace: a React web UI with a FastAPI backend that provisions per-workspace sandboxes (Docker or host), runs Claude Code via claude-agent-sdk/Claude Code CLI, and routes requests to Anthropic directly or through an Anthropic-compatible bridge for other providers (OpenAI/OpenRouter/GitHub Copilot/custom compatible endpoints). It also supports extension surfaces like MCP servers, skills/agents/commands, and streaming with SSE and cancellation.

Evaluated Mar 30, 2026 (0d ago)

Repo ↗ DevTools ai-ml devtools automation infrastructure sandbox agent claude-code fastapi react typescript mcp

⚙ Agent Friendliness

/ 100

Can an agent use this?

🔒 Security

/ 100

Is it safe for agents?

⚡ Reliability

/ 100

Does it work consistently?

Score Breakdown

⚙ Agent Friendliness

MCP Quality

Documentation

Error Messages

Auth Simplicity

Rate Limits

🔒 Security

TLS Enforcement

Auth Strength

Scope Granularity

Dep. Hygiene

Secret Handling

Security posture can’t be fully verified from README. The app requires a SECRET_KEY for JWT signing, which is a positive signal, but scope/authorization granularity and safe secret handling practices (e.g., no logging of tokens) are not documented. Sandboxed execution via Docker/host helps isolate code execution, but the deployment hardening, network isolation, and credential storage practices are not specified. TLS enforcement for API is not explicitly stated (assumed via typical deployment), so score is moderate rather than definitive.

⚡ Reliability

Uptime/SLA

Version Stability

Breaking Changes

Error Recovery

Best When

You want to run Claude Code capabilities on your own infrastructure (web IDE + sandboxed execution) and need provider switching/routing while keeping consistent workspace state and tool/MCP orchestration.

Avoid When

You cannot secure JWT/API endpoints and sandbox/runtime secrets, or you require a stable API contract without anticipating breaking changes (the project states it is under active development).

Use Cases

• Self-hosted web IDE for Claude Code-style agent workflows with sandboxed execution
• Multi-provider model routing while keeping a shared workspace filesystem/context
• Team/project workspace management with isolated sandboxes per project
• Running Claude Code tools in Docker/host isolation for safer code execution
• Extending workflows via MCP servers/skills/commands/plugins
• Web-based operation with health/readyz endpoints and admin UI

Not For

• Compliance-first environments that require formally documented security controls, audit logs, and hardened deployment guidance without review
• Situations needing a lightweight, single-purpose API (this is a full application stack)
• Users who cannot manage Docker/Compose or local sidecar/container runtime dependencies
• Environments where provider credentials cannot be stored/handled by a self-hosted service without additional controls

Interface

REST API

Yes

GraphQL

gRPC

MCP Server

SDK

Webhooks

OpenAPI Spec ↗

Authentication

Methods: JWT (SECRET_KEY used for JWT signing per .env) Provider tokens in UI/settings (Anthropic token; OpenAI device flow; GitHub Copilot device flow; OpenRouter API key; custom Anthropic-compatible base_url + token) Basic use via browser session (implied by web UI; exact auth flow not specified in README)

OAuth: Yes Scopes: No

Authentication for the app API/UI appears to rely on JWT signing using SECRET_KEY. Provider authentication varies by provider: Anthropic token directly; OpenAI and Copilot use device-flow in UI; OpenRouter uses API key; custom uses base_url+token. README does not document JWT claim structure or fine-grained scope model for API authorization.

Pricing

Free tier: No

Requires CC: No

Self-hosted open-source (Apache 2.0) per README; cost depends on your infrastructure and upstream provider usage.

Agent Metadata

Pagination

none

Idempotent

False

Retry Guidance

Not documented

Known Gotchas

⚠ Breaking changes are expected because the project is under active development.
⚠ Provider routing and sandbox execution introduce operational complexity; agent workflows may fail if sandbox runtime or bridge services are misconfigured.
⚠ Per-workspace sandbox provider is locked at creation time; switching may require recreating workspaces.
⚠ README mentions streaming SSE with cancellation and resumable events, but does not specify agent retry/resumption semantics or idempotency guarantees.

Alternatives

Open-WebUI-style wrappers for model providers (if you don’t need Claude Code CLI/sandbox harness) Self-hosted Claude Code alternatives/variants that integrate a code-execution sandbox directly Other agent orchestration platforms with sandboxing (e.g., generic agent frameworks) Standalone Claude Code CLI with your own local web front-end

Full Evaluation Report

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

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

Score Monitoring

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

Continuous monitoring

$3/mo

API endpoint ↗ Agent guide ↗ Report inaccuracy

Scores are editorial opinions as of 2026-03-30.