Gemini CLI

Coding · tested 2026-08-06 · re-test due 2026-11-04 · by the Hlido desk, not the vendor

In short: Google's terminal coding agent, backed by the strongest distribution in the category — and carrying a prominent replacement notice for free-tier users.

3 PASS · 2 FAIL of 5 public-surface claims

Quick answer

Gemini CLI scores 76/100 (STEADY) on Hlido’s independent, hands-on test (reviewed 2026-08-06). STEADY (76) because it is a first-party Google tool with open development, standard npm distribution and the strongest model access and distribution in its category. Pricing: Free tier (free entry point documented).

Gemini CLI brings Google's models into the terminal for the now-standard agentic coding loop: query and edit large codebases, generate applications from images or PDFs, and automate multi-step workflows. It installs as a global npm package, is developed in the open under the google-gemini organisation, and inherits the two advantages no independent competitor can match — first-party model access and Google's distribution. For a developer already inside the Google ecosystem, that combination is hard to argue with. The finding that dominates this review is at the top of the vendor's own page: a site-wide banner stating that for unpaid-tier and Google One users, Gemini CLI will be replaced by Antigravity CLI. Whatever the migration details turn out to be, a vendor announcing a replacement product for its free tier is material to anyone weighing adoption, and it belongs in the first paragraph of any honest assessment rather than a footnote. We are reporting the notice as displayed; we did not verify the transition timeline, and readers evaluating this should check the linked blog post for current terms. The second caveat is about this review's own confidence rather than the product. The captured surface was thin — a marketing homepage, a cookie notice and a navigation bar — so the substance sits behind documentation and the repository, neither of which this public-surface run exercised. The score reflects a strong, well-distributed, openly developed tool assessed on a shallow surface with an unresolved continuity question, and should be read with that confidence level in mind.

Why STEADY

STEADY (76) because it is a first-party Google tool with open development, standard npm distribution and the strongest model access and distribution in its category. Held below VITAL by a vendor-published notice that free-tier users will be moved to a replacement product — an unresolved continuity signal — and by a thin captured surface that left the substantive capability claims unexercised. Confidence is low-medium for that reason.

Public-surface checklist

What it does well

What it fails at

Best for

  • Developers already committed to Google Cloud or the Gemini model family
  • Terminal-first engineers who want a first-party agentic coding tool rather than a third-party wrapper
  • Teams comfortable tracking a fast-moving vendor roadmap, including product replacements
  • Multimodal workflows that start from images or PDFs rather than text alone

Not recommended for

  • Free-tier users seeking long-term stability, given the published replacement notice
  • Teams needing model portability across vendors — this is single-vendor by design
  • Buyers requiring published benchmarks before adopting a coding agent
  • Anyone avoiding Google account dependencies on principle or policy

Pricing & access

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 2026-08-05.

Compared to

Agent relevance

CLI

Agentic-Commerce Readiness 38/100 · SURFACE-ONLY

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

Terminal CLI installed globally via npm and driven by command invocation; an extensions surface is advertised. It is an agent a developer runs, not a service other agents call — there is no documented programmatic endpoint for agent-to-agent use on the captured surface.

Agent-friendly score: 5/10

Evidence

scorecard.json · 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 2026.08 · Next review due 2026-11-04

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

Hlido trust score

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/google-gemini-gemini-cli.svg)](https://hlido.eu/check/?agent=google-gemini-gemini-cli)

HTML

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