Skip to content
ADE Index

capstan vs Claude Code CLI vs Gemini CLI vs Hermes Agent

Up to four tools, row by row. We do not guess: a cell we have not checked says so.

This page compares the four CLI agents only. Claude Code has 4 other listings and Gemini CLI has 1 other listing, each with its own page.

How they differ

Pick this if…

Paraphrased from each maker’s own description. Select a line for the quote.

Checked for Claude Code CLI, Gemini CLI and Hermes Agent, not yet for capstan: Price (Claude Code CLI: ; Gemini CLI: ; Hermes Agent: ), Bring your own key (Claude Code CLI: ; Gemini CLI: ; Hermes Agent: )

Checked for capstan, Claude Code CLI and Gemini CLI, not yet for Hermes Agent: Systems (capstan: , , ; Claude Code CLI: , , , ; Gemini CLI: , , )

Not yet checked for any of these tools: agents it runs.

Dotted values open their quote.

What each can do

Select an answer to see the maker’s sentence, or what we searched.

We have not checked capstan’s capabilities yet. 31 capabilities are checked for at least two of these tools. 3 get different answers: Editor integration, Review diffs and Browser automation; 10 differ only in a limit. 6 others are checked for one tool only.

Differences only13
  • Yesthe maker documents it
  • With add-onAdd-onsomething the maker publishes that you install or run
  • *its limit or add-on is named under the row
  • Emptynot yet checked
  • From the listingthe listing’s own facts; not one of the capabilities we check

Models, keys and usage

Named model providersFrom the listing
,
,,,
Listing also names: Microsoft Foundry (Azure AI Foundry)
,,,,,,,,
+30 more model providers
,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
Listing also names 11 othersAmazon Bedrock, SpaceXSI (Grok), Moonshot AI, Vercel AI Gateway, Alibaba (Qwen), OpenCode Console (OpenCode Zen), Microsoft Foundry (Azure AI Foundry), Meta AI (Llama), Nebius Token Factory (Nebius AI Studio), Novita AI, Command Code
Local modelsNot yet checked: capstan, Claude Code CLI, Gemini CLI
Not yet checked for capstan
Not yet checked for Claude Code CLI
Not yet checked for Gemini CLI
Custom OpenAI-compatible endpointNot yet checked: capstan, Claude Code CLI, Gemini CLI
Not yet checked for capstan
Not yet checked for Claude Code CLI
Not yet checked for Gemini CLI
Token and cost usageNot yet checked: capstan
Not yet checked for capstan
Token view for API users

MCP, hooks, skills and subagents

MCP serversNot yet checked: capstan
Not yet checked for capstan
local and remote
local and remote
local and remote
HooksNot yet checked: capstan
Not yet checked for capstan
SkillsNot yet checked: capstan
Not yet checked for capstan
SubagentsNot yet checked: capstan
Not yet checked for capstan
PluginsNot yet checked: capstan
Not yet checked for capstan
Custom commandsNot yet checked: capstan
Not yet checked for capstan

Editors, scripts and CI

Editor integrationNot yet checked: capstan
Not yet checked for capstan
VS Code extension or JetBrains plugin
Gemini CLI Companion extension
ACP editors
Headless modeNot yet checked: capstan
Not yet checked for capstan
JSON outputNot yet checked: capstan
Not yet checked for capstan
GitHub integrationNot yet checked: capstan, Gemini CLI, Hermes Agent
Not yet checked for capstan
Claude Code GitHub Action
Not yet checked for Gemini CLI
Not yet checked for Hermes Agent
Commit attribution controlNot yet checked: capstan, Gemini CLI, Hermes Agent
Not yet checked for capstan
Not yet checked for Gemini CLI
Not yet checked for Hermes Agent

Review, steer and undo

Plan modeNot yet checked: capstan, Hermes Agent
Not yet checked for capstan
Not yet checked for Hermes Agent
Read-only modeNot yet checked: capstan, Hermes Agent
Not yet checked for capstan
Through plan mode
Not yet checked for Hermes Agent
Review diffsNot yet checked: capstan
Not yet checked for capstan
Gemini CLI Companion extension
Needs checkpoints on
Undo file editsNot yet checked: capstan
Not yet checked for capstan
Opt-in; local and SSH only

Parallel agents

Parallel sessionsNot yet checked: capstan, Gemini CLI
Not yet checked for capstan
Not yet checked for Gemini CLI
Git worktreesNot yet checked: capstan
Not yet checked for capstan
Experimental; turn on first

Context and memory

Project instruction filesNot yet checked: capstan
Not yet checked for capstan
Memory across sessionsNot yet checked: capstan
Not yet checked for capstan
Resume a sessionNot yet checked: capstan
Not yet checked for capstan
Browse past sessionsNot yet checked: capstan
Not yet checked for capstan

Remote and mobile control

Remote controlNot yet checked: capstan, Gemini CLI
Not yet checked for capstan
Needs a Claude plan, not an API key
Not yet checked for Gemini CLI
Steer from a phoneNot yet checked: capstan, Gemini CLI
Not yet checked for capstan
Needs a Claude plan, not an API key
Not yet checked for Gemini CLI
Unblocks Kanban workers
NotificationsNot yet checked: capstan
Not yet checked for capstan
Depends on the terminal
Experimental; off by default

Sandbox, permissions and privacy

OS-level sandboxNot yet checked: capstan, Hermes Agent
Not yet checked for capstan
Not on native Windows
Opt-in; Seatbelt is macOS only
Not yet checked for Hermes Agent
Approval modesNot yet checked: capstan
Not yet checked for capstan
Runs in a containerNot yet checked: capstan, Claude Code CLI
Not yet checked for capstan
Not yet checked for Claude Code CLI
Opt-in Docker or Podman sandbox
Official Docker setup
Telemetry opt-outNot yet checked: capstan, Gemini CLI, Hermes Agent
Not yet checked for capstan
Not yet checked for Gemini CLI
Not yet checked for Hermes Agent

Images, web and your computer

Image inputNot yet checked: capstan
Not yet checked for capstan
Fetch URLsNot yet checked: capstan, Hermes Agent
Not yet checked for capstan
Not yet checked for Hermes Agent
Browser automationNot yet checked: capstan, Gemini CLI
Not yet checked for capstan
Claude in Chrome extension
Not yet checked for Gemini CLI
Computer useNot yet checked: capstan, Gemini CLI, Hermes Agent
Not yet checked for capstan
macOS only; Pro, Max or Ultra plan
Not yet checked for Gemini CLI
Not yet checked for Hermes Agent
Voice inputNot yet checked: capstan, Gemini CLI
Not yet checked for capstan
Claude.ai sign-in; no cloud or SSH
Not yet checked for Gemini CLI
Not yet checked: , , ,

Each answer is about the CLI agent in its column. Something the maker publishes separately, such as an editor plugin or a GitHub Action, shows as “With add-on”, even when it has its own listing here. How capabilities are checked

About these tools

capstan

Claude Code CLI

Gemini CLI

Hermes Agent

How to read the Index score

At a glance

Index score out of 100

capstanNot enough data yetClaude Code CLI92Gemini CLI44Hermes Agent48

Mindshare HN mentions, 90 days

capstan0Claude Code CLI4,602Gemini CLI58Hermes Agent84

Adoption GitHub stars, not scored

capstan17Claude Code CLI156kGemini CLI109kHermes Agent264k

Release days last 90 days

capstan5Claude Code CLI57Gemini CLI12Hermes Agent19

Lowest monthly plan billed monthly, lower is cheaper

capstanNot yet checkedClaude Code CLI$20/moGemini CLINot yet checkedHermes AgentNot yet checked

Platforms systems and surfaces

capstan3Claude Code CLI5Gemini CLI3Hermes AgentNot yet checked

Votes can only raise a score: up to 20% of it reflects upvotes, graded on a curve against the most-voted tool. The square-root curve has a floor of 100 effective votes. Votes decay with a 90-day half-life. Maker self-votes do not count. Counts below 10 are hidden for display only. After activation, the score is 80% mindshare plus 20% of the greater of mindshare and the vote score. GitHub stars do not affect the Index score. Mindshare uses a log scale; votes use the square-root curve and a 100-vote floor. Ranks use unrounded scores. Mindshare and adoption bars are log-scaled; the others are linear within this comparison. How the score is built

Differences only
4 of 4
capstantheStrangeAdventurer · CLI agent
Claude Code CLIAnthropic · CLI agent
Gemini CLIgoogle-gemini · CLI agent
Hermes AgentCLI agent
Interface
Terminal session listing file searches and reads while tests run
Terminal session after an MCP tool call generated three image files
What it is
A lightweight, extensible terminal coding agent built in C with an embedded Lua runtime.
Terminal coding tool that executes routine tasks and handles git workflows
Open-source terminal agent for enterprise Gemini Code Assist workflows on Google Cloud
Terminal agent for writing, reviewing and running code in local projects
Category
CLI agent
CLI agentClaude Code family
CLI agentGemini CLI family
CLI agent
Pricing
Starts at (billing shown)
Not yet checked
Lowest monthly plan
Not yet checked
Not yet checked
Not yet checked
Free tier
Not yet checked
Not yet checked
Not yet checked
Plans
Not yet checked
Not yet checked
Bring your own key
Not yet checked
Yes
Yes
Yes
Runs on
Linux
Yes
Yes
Yes
Not yet checked
Terminal
Yes
Yes
Yes
Not yet checked
Web
Not yet checked
Yes
Not yet checked
Not yet checked
Windows
Not yet checked
Yes
Not yet checked
Not yet checked
macOS
Yes
Yes
Yes
Not yet checked
Models and agents
Drives these agents
Not yet checked
Not yet checked
Not yet checked
Not yet checked
Source and activity
Source
GitHub stars
Most
Release days, last 90 days
5
57Most
12
19
Release days per week
Latest release, any channel
v0.6.0 22 Oct 2026
v2.1.327 17 Nov 2026
v0.68.0 10 Nov 2026
Index score
Score and rank
Not enough data yet
92 #1 Score 91.6; ranked on the unrounded score
44 #18 Score 44.3; ranked on the unrounded score
48 #14 Score 48.2; ranked on the unrounded score
Mindshare
00 Hacker News mentions in the last 90 days
924,602 Hacker News mentions in the last 90 days
4458 Hacker News mentions in the last 90 days
4884 Hacker News mentions in the last 90 days
Adoption, not scored
2417 stars, read 18 Nov 2026
98156,056 stars, read 18 Nov 2026
95108,907 stars, read 18 Nov 2026
100264,230 stars, read 18 Nov 2026
Votes
—Votes can only raise a score: up to 20% of it reflects upvotes, graded on a curve against the most-voted tool. The square-root curve has a floor of 100 effective votes. Votes decay with a 90-day half-life. Maker self-votes do not count. Counts below 10 are hidden for display only.
—Votes can only raise a score: up to 20% of it reflects upvotes, graded on a curve against the most-voted tool. The square-root curve has a floor of 100 effective votes. Votes decay with a 90-day half-life. Maker self-votes do not count. Counts below 10 are hidden for display only.
—Votes can only raise a score: up to 20% of it reflects upvotes, graded on a curve against the most-voted tool. The square-root curve has a floor of 100 effective votes. Votes decay with a 90-day half-life. Maker self-votes do not count. Counts below 10 are hidden for display only.
—Votes can only raise a score: up to 20% of it reflects upvotes, graded on a curve against the most-voted tool. The square-root curve has a floor of 100 effective votes. Votes decay with a 90-day half-life. Maker self-votes do not count. Counts below 10 are hidden for display only.
Rank history
Not ranked
First ranked 18 Nov 2026
First ranked 18 Nov 2026
First ranked 18 Nov 2026
Full page

Watch capstan vs Claude Code CLI vs Gemini CLI vs Hermes Agent on YouTube

More comparisons

Build your own