devir
Devir is a Go-based developer utility/CLI that runs and manages multiple dev services (processes) in a project, displays their logs in a Bubble Tea terminal UI with filtering/search and colored output, and provides integration surfaces including an MCP server (for Claude Code) and a Chrome DevTools WebSocket bridge. It also supports a daemon mode (Unix socket) so multiple clients (TUI and MCP) can share the same running services and log stream, plus port detection/kill helpers.
Score Breakdown
⚙ Agent Friendliness
🔒 Security
Security is not well-specified in the README. While TLS is likely not applicable to the Unix socket MCP transport, the WebSocket bridge (port 9222) is not documented as secured (no auth/TLS mention). The system can kill processes on ports and control service execution, which elevates risk if exposed to untrusted users on the same host. Dependency hygiene and secret-handling practices are not described; score reflects uncertainty rather than evidence.
⚡ Reliability
Best When
You want a local developer workflow to coordinate multiple service processes with shared logs, and you want an MCP interface for AI-assisted control inside your own machine/project environment.
Avoid When
You are operating in a multi-tenant or untrusted environment where local daemon/MCP access could be abused, or you need strong guarantees around authentication, authorization, and secure transport for programmatic control surfaces.
Use Cases
- • Run multiple local dev services from a single config (devir.yaml)
- • Monitor and filter per-service logs in an interactive terminal UI
- • Use MCP tools to start/stop/restart services and fetch logs/status programmatically (e.g., from Claude Code)
- • Share one running service set across multiple client instances (daemon mode)
- • Use Chrome DevTools panel to view the same logs via WebSocket
- • Health checks and status signaling via a .devir-status file
- • Mitigate port conflicts by detecting and killing processes bound to ports
Not For
- • Production-grade process supervision for sensitive/hosted workloads
- • Security-audited, least-privilege remote control over networks without additional hardening/auth
- • Teams needing documented REST/OpenAPI/SDK contracts
- • Use cases requiring strong data governance guarantees for logs over external channels (no guarantees stated)
Interface
Authentication
The README documents MCP server tools and a daemon using a Unix socket per project directory, but does not describe any authentication/authorization mechanism for MCP tool invocation or WebSocket access.
Pricing
No pricing information is provided (open-source CLI context implied by MIT license).
Agent Metadata
Known Gotchas
- ⚠ Local-only nature: daemon socket is per project directory; agents must ensure they run in the correct cwd and target the same project instance.
- ⚠ Statefulness: start/stop/restart and port-kill actions are side-effectful; repeated tool calls may cause unintended restarts or process termination.
- ⚠ Real-time log streaming: agents relying on log availability should handle timing/race conditions after service restarts.
- ⚠ Clipboard and UI actions are not relevant for agent tool usage; ensure agent uses MCP tools rather than UI keystrokes.
Alternatives
Full Evaluation Report
Comprehensive deep-dive: security analysis, reliability audit, agent experience review, cost modeling, competitive positioning, and improvement roadmap for devir.
AI-powered analysis · PDF + markdown · Delivered within 30 minutes
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
Scores are editorial opinions as of 2026-03-30.