Gemini CLI vs Claude Code
Updated : the last week a score, rank or safety result on this page changed.
- Maintenance 30%7177
- Adoption 30%6982
- Docs 20%6920
- Community 20%, not measured yet––
Gemini CLI scores 5 points higher. Gemini CLI leads on docs; Claude Code leads on maintenance and adoption.
What this does not tell you
Scores are percentiles of public GitHub signals. Nobody here tested these tools hands-on, so the higher score is not a verdict on which suits your project. Yellow marks the higher value.
Which one fits you
From measured data and licences, not from testing.
Pick Gemini CLI if
- You rely on thorough README docs (docs 69 vs 20).
- You need a permissive open-source licence (Apache-2.0).
Pick Claude Code if
- Recent activity and releases matter most to you (maintenance 77).
- You want the more widely adopted option (adoption 82).
Safety side by side
Automated scan, 2 Nov 2026. Not a manual audit.
- CautionGemini CLI9 critical-severity known vulnerabilities in runtime dependencies (OSV). Agent file pipes a script into a shell
- No dependencies scannedClaude CodeCode checks clean, no lockfile to check
Why different labels?
A label partly reflects how much of each tool is visible and scannable, not only how risky it is. A tool without a lockfile cannot get a dependency check.
- Gemini CLI: 9 critical-severity known vulnerabilities in runtime dependencies (OSV); Skill/agent file instructs piping a remote script into a shell: .gemini/skills/agent-tui/SKILL.md — curl -fsSL https://raw.githubusercontent.com/pproenca/agent-tui/master/install.sh | sh
- Claude Code: code and instruction-file checks found nothing, but there is no runtime lockfile or manifest, so dependencies were not checked
Repo facts
Yellow = the higher value, where higher is better.
| Fact | Gemini CLI | Claude Code |
|---|---|---|
| Score | 70 | 65 |
| Stars | 108,661 | 154,199 |
| Last push | 2 Nov | 1 Nov |
| Releases, 90 days | 10+ | 10+ |
| Licence | Apache-2.0 | Proprietary (Anthropic Commercial Terms) |
| Open source | Open source | Proprietary |
| Language | TypeScript | TypeScript |
| Type | AI tool | AI tool |
| Runs locally | yes | yes |
| Hosted option | unknown | mentioned |
“Runs locally” and “Hosted option” are read from each README automatically (or from the official docs where noted); “unknown” means no evidence was found, not that it is absent.
Short answer
Gemini CLI scores 70 and Claude Code 65 this week. Gemini CLI leads on README docs (69 vs 20), Claude Code leads on adoption (82 vs 69) and Claude Code leads on maintenance (77 vs 71). Gemini CLI is Apache-2.0, open source; Claude Code is Proprietary (Anthropic Commercial Terms).
Read the full answer
Safety, Gemini CLI: the automated scan gives it a Caution label for 9 critical-severity known vulnerabilities in runtime dependencies (OSV) and a skill or agent instruction file (.gemini/skills/agent-tui/SKILL.md) that tells the agent to pipe a remote script into a shell. Claude Code: the scan found no issues in the code and agent instruction files of the public repo, but there was no dependency list to check: osv-scanner found no runtime lockfile or package manifest, so the label is “No dependencies scanned”, not a passed dependency check. So a different label partly reflects how much of each tool is visible and scannable, not only how risky it is. Gemini CLI safety details · Claude Code safety details
Heads-up from Google's docs: “Unpaid tier and Google One users: Gemini CLI was replaced by Antigravity CLI on June 18th, 2026.” Check which product your account gets. (geminicli.com, checked 27 Oct 2026)
Sandbox and permission options (from each vendor's docs)
- Gemini CLI has optional sandboxing:
gemini -s,GEMINI_SANDBOX=true|docker|podman|sandbox-exec|runsc|lxc, or"tools": {"sandbox": "docker"}insettings.json. Google's own note: “Sandboxing reduces but doesn't eliminate all risks.” - Claude Code in Manual mode starts read-only and asks before edits and commands; bash can be sandboxed with
/sandbox; organizations can enforce settings centrally.
Sources: Gemini CLI docs, sandboxing · Claude Code security docs.
Which should you pick?
- Open-source license: Gemini CLI, ideally with sandboxing on and after reviewing the flagged skill file.
- Already on Anthropic plans, want managed enterprise controls: Claude Code.
- More: OpenCode vs Claude Code · OpenCode vs Gemini CLI · Gemini CLI vs OpenHands · all coding agents
Questions about Gemini CLI vs Claude Code
Answers use this week's StackGrade data.
Gemini CLI or Claude Code: which scores higher on StackGrade?
Gemini CLI scores 5 points higher in Week 45: Gemini CLI 70, Claude Code 65 out of 100. The score combines maintenance, adoption and docs (percentiles of public GitHub signals): Gemini CLI 71, 69 and 69; Claude Code 77, 82 and 20. Community (20% weight) is not measured yet. It is not a hands-on test, so a higher score does not say which tool suits your project.
Is Gemini CLI or Claude Code safer to use?
Safety labels from the automated scan of 2 Nov 2026: Gemini CLI: Caution, 9 critical-severity known vulnerabilities in runtime dependencies (OSV); Claude Code: No dependencies scanned, code checks clean, no lockfile to check. “No dependencies scanned” means the code checks were clean but there is no lockfile, so dependencies could not be checked; it is not a clean bill. It is an automated scan, not a manual audit.
Which is more widely used, Gemini CLI or Claude Code?
By GitHub stars: Claude Code 154,199, Gemini CLI 108,661. Adoption score (percentile among all 52 tracked tools): Gemini CLI 69, Claude Code 82. Stars are one public signal and do not measure real-world use.
Which licences do Gemini CLI and Claude Code use?
Gemini CLI: Apache-2.0; Claude Code: Proprietary (Anthropic Commercial Terms). Read each licence text before commercial use.
Should I pick Gemini CLI or Claude Code?
Pick Gemini CLI if you rely on thorough README docs (docs 69 vs 20); you need a permissive open-source licence (Apache-2.0). Pick Claude Code if recent activity and releases matter most to you (maintenance 77); you want the more widely adopted option (adoption 82). This comes from measured public data and licences only, not from hands-on testing.
Full comparison table, every criterion
| Gemini CLIgoogle-gemini/gemini-cli | Claude Codeanthropics/claude-code | |
|---|---|---|
| Score | 70 | 65 |
| Rank | #4 overall · #1 in use case | #11 overall · #4 in use case |
| Maintenance (30%) | 71 | 77 |
| Adoption (30%) | 69 | 82 |
| Docs (20%) | 69 | 20 |
| Community (20%) | not measured | not measured |
| Safety | Caution
| No dependencies scanned |
| License | Apache-2.0 | Proprietary (Anthropic Commercial Terms) |
| Open source | Open source | Proprietary |
| Language | TypeScript | TypeScript |
| Stars | 108,661 | 154,199 |
| Last push | ||
| Type | AI tool | AI tool |
| Runs locally | yes (README: "npm install -g") | yes (README: "npm install -g") |
| Hosted option | unknown | mentioned (official docs: code.claude.com/docs/en/claude-code-on-the-web) |
Get the board every Monday
One email a week: the new top 10, the biggest score changes and every safety label that changed.
Free. Double opt-in: you confirm by email first, unsubscribe in one click.