What did Hlido score SquadCue?

SquadCue scored 73/100 (STEADY) in Hlido's independent, hands-on review.

How much does SquadCue cost?

Hlido recorded the pricing model as: Open source. Derived from evidence Hlido holds, not vendor-supplied; verify current prices on the vendor's pricing page.

Does any vendor pay Hlido for placement?

No. Hlido takes no money from the agents it rates — scoring weights stay private and the evidence behind every verdict is public.

SquadCue

Name: SquadCue review
Item: SquadCue
Rating: 73
Author: Hlido Editor

Workflow & Automation · tested 2026-08-19 · re-test due 2026-11-19 · by the Hlido desk, not the vendor

In short: A local cockpit that turns your Claude Code sessions into named 'employees' you supervise through an approval inbox — early and single-user, but unusually honest about exactly what it is and isn't.

Visit SquadCue → scorecard.json

Quick answer

SquadCue scores 73/100 (STEADY) on Hlido’s independent, hands-on test (reviewed 2026-08-19). STEADY (73) for an MIT-licensed, genuinely useful local cockpit (session-as-employee, canvas pipelines, a real human-in-the-loop approval inbox) whose standout is exceptional honesty about its own limits and non-security Pricing: Open source (free entry point documented).

Hlido Editor · 2026-08-19

SquadCue (MIT, v0.1.2) is a laptop-scale control plane for AI CLI sessions. Recent Claude Code sessions on your machine become named contacts you chat with, resume across restarts, and supervise from your phone; a visual Drawflow canvas orchestrates repeatable pipelines on top. Its centrepiece is an approval inbox: with tools allowed, a PreToolUse hook routes risky actions (shell, file writes, network fetches) into a web + Telegram queue where the first response wins and a timeout means deny. What earns Hlido's respect is the honesty. The project states plainly and repeatedly that the inbox is a supervision workflow for a trusted local setup, not a security boundary, points to a SECURITY.md that spells out what it cannot stop, and documents its opt-outs, its loopback-only enforcement, and a production 'learned the hard way' caveat that background jobs spawned in a chat turn may die with it. That candour is rare and valuable. The trade-off is maturity and reach: it is explicitly single-user by design, Windows is the daily-driver platform (canvas shell nodes run via PowerShell; session relaunch is Windows-only today), Claude Code is the only first-class engine (Codex is a red-team add-on, other CLIs a roadmap adapter), and the whole thing was extracted in July 2026 from one person's tooling. The 'immortal employees' durable-memory model is roadmap, not shipped. Reviewed from the project surface only; Hlido did not run the cockpit.

Why STEADY

STEADY (73) for an MIT-licensed, genuinely useful local cockpit (session-as-employee, canvas pipelines, a real human-in-the-loop approval inbox) whose standout is exceptional honesty about its own limits and non-security-boundary status — held at low-medium confidence by early-stage maturity (v0.1.2, single-user by design, Windows-primary, Claude-Code-first) and a surface-only review. The candour raises trust; the maturity caps the score.

What we saw

4 screenshots captured by the Hlido engine during the reviewed run (run-22cf84e4e27007f0-squadcue-com). Our own captures — not vendor marketing material.

SquadCue — run screenshot 1 (home.png) — home.png

SquadCue — run screenshot 2 (page_.png) — page_.png

SquadCue — run screenshot 3 (page__why-squadcue.png) — page__why-squadcue.png

SquadCue — run screenshot 4 (page__features.png) — page__features.png

What it does well

Turns real Claude Code sessions into named, resumable 'employees' you chat with and supervise — a concrete, useful local abstraction
A real human-in-the-loop approval inbox (web + Telegram, first-response-wins, timeout = deny) with recorded, auditable decisions
Exceptionally honest self-documentation: it states the inbox is not a security boundary, links a SECURITY.md of what it cannot stop, and flags hard-won production caveats
Sensible local security posture — loopback-only enforced at startup, CSRF/DNS-rebind guards, path containment, SSRF guard on fetch, no shell template interpolation
Zero-infra install (pip install squadcue; plain JSON/JSONL + SQLite state you can grep); MIT-licensed

What it fails at

Early stage — v0.1.2, extracted from one person's tooling; durable cross-session 'immortal employee' memory is roadmap, not shipped
Single-user by design (loopback only; no auth/RBAC until a second user exists)
Windows is the daily-driver platform — canvas shell nodes run via PowerShell and session relaunch is Windows-only today; Linux/macOS need adaptation
Claude Code is the only first-class engine; Codex is a red-team add-on and other CLIs are a roadmap adapter layer
Surface-only review — Hlido did not run the cockpit, so behaviour is unverified

Red flags

By the project's own statement, the approval inbox is a supervision workflow, not a security boundary — approving a shell command approves the whole program it runs; do not treat it as a sandbox

Best for

Solo operators running Claude Code (and optionally Codex) all day who want a local cockpit with named sessions and phone-based supervision
Users who want a real approval gate over risky agent actions and value plain, greppable local state
Windows-primary developers comfortable with an early, honestly-scoped open-source tool

Not recommended for

Teams needing multi-user access, auth or RBAC
Anyone expecting the approval inbox to be a security sandbox — it is explicitly not
Linux/macOS users wanting full parity today, or users of CLIs other than Claude Code/Codex

Pricing & access

ModelOpen source
Free entry pointYes — a free tier or open-source edition is documented

Derived from Hlido-held evidence only (engine checklist + editorial text); quotes are verbatim from the scorecard; not vendor-supplied; re-derived daily. Verify current prices on the vendor's pricing page. Last verified 2026-08-18.

Related agents

Agent relevance

API CLI Behavioral-testable

Agentic-Commerce Readiness 48/100 · SURFACE-ONLY

Independent readiness for agent delegation & transaction. How it’s scored · check live

Install via pip and run locally (FastAPI server + single-file cockpit); it discovers Claude Code sessions as 'employees', routes risky tool calls through an approval inbox (web + Telegram), and orchestrates pipelines on a canvas. Control plane and state stay on your machine.

Agent-friendly score: 7/10

Evidence

Turns real Claude Code sessions into named, resumable 'employees' you chat with and supervise — a concrete, useful local abstraction — source (2026-08-19) verified
A real human-in-the-loop approval inbox (web + Telegram, first-response-wins, timeout = deny) with recorded, auditable decisions — source (2026-08-19) verified
Exceptionally honest self-documentation: it states the inbox is not a security boundary, links a SECURITY.md of what it cannot stop, and flags hard-won production caveats — source (2026-08-19) verified
Sensible local security posture — loopback-only enforced at startup, CSRF/DNS-rebind guards, path containment, SSRF guard on fetch, no shell template interpolation — source (2026-08-19) verified
Hands-on runtime behaviour (executing the tool / a live task) — source (2026-08-19)

scorecard.json · transparency passport · registry · methodology

More: compare agents · best of · developer tools · incident registry

Verdict by Hlido Editor, our automated editorial system · Method: public-surface-tier-2+editorial-narrative-v2 · Methodology version 2026.05 · Next review due 2026-11-19

How this page was produced. The scores, claim verdicts and evidence come from automated hands-on testing of the product’s public surface. The written analysis is drafted by an AI system, and pages publish without a person reviewing each one. Hlido publishes this record and answers for it — tell us if anything here is wrong and we will correct it.

Embed this trust badge

Live, always-current independent score — free to embed on your site or README. No vendor pays for placement.

Markdown

[![Hlido trust score](https://hlido.eu/badge/squadcue.svg)](https://hlido.eu/check/?agent=squadcue)

HTML

<a href="https://hlido.eu/check/?agent=squadcue"><img src="https://hlido.eu/badge/squadcue.svg" alt="Hlido trust score"></a>