AgentMesh.help
Freemium
Developers integrating AI agents into task workflows
Goose
FreeOpen SourceAPISelf-HostedAgentic
Goose is an open-source AI agent you run as a desktop app, command line or API. Connect a model and MCP extensions, and it can edit files, run commands and work with other services.
Similar tools
Claude CodePaidAnthropic's coding agent for editing repositories, running commands and tests, and reviewing…
WorkBuddyFreemiumWorkBuddy runs as a local-first agent on the desktop, autonomously chaining file access, web…
Hermes DesktopFreeHermes Studio is an open-source, self-hosted dashboard that wraps Hermes Agent in a control…
ChatGPTFreemiumOpenAI's assistant for writing, research, analysis and file creation through Chat, Work and…
Get This Tool
License: Apache-2.0 Any use incl. commercial
Local-run terms: Users can download, install, modify, and run the desktop app, CLI, or API locally under Apache 2.0 terms.
Pricing
- Model
- Free
Summary
We ran Goose 1.54.0 on three small jobs: create a file, summarize a report, and count the rows in a CSV with a shell command, three runs each in fresh containers. With a small local model (qwen2.5:3b on 3 CPU cores) it passed 5 of 9, the best small-model score of the three agents we tested. Every failure was the model's mistake, such as counting the CSV header row, and Goose itself exited cleanly on every run. With Claude Opus 5.5 it passed all 9 and was the cheapest agent to run: about 7,800 tokens per task and $0.14 for all nine runs, against $0.19 for OpenCode and $0.27 for Hermes Agent. It was also the fastest on two of the three tasks. Recipes let you save a task's instructions and extension settings to reuse or share.
Bottom line: Of the three agents we tested, Goose did the most with a small local model and used the fewest tokens with Claude Opus 5.5. A strong pick for file and shell tasks; just check where each extension and model provider sends your data.
We installed and ran it
- How we installed it
- Vendor one-line install script
- Docker image size on disk
- 0.71 GB
- Local model we used
- qwen2.5:3b via Ollama
- Agent tasks passed with the local model
- 5 of 9
- Fixed prompt sent with every model call
- ~1,150 tokens
- Outside hosts it tried to reach (we blocked them)
- models.dev
| Task (qwen2.5:3b) | Passed | Median time |
|---|---|---|
| Create a file with given text | 3 / 3 | 10.5 s |
| Read a report and write a 2-sentence summary | 1 / 3 | 33.1 s |
| Count the data rows in a CSV with a shell command | 1 / 3 | 15.1 s |
With Claude Opus 5.5 (Anthropic API)
- Model
- Claude Opus 5.5 (claude-opus-5-5)
- Agent tasks passed
- 9 of 9
- API cost for all 9 runs (list price)
- $0.14
- Tokens per task (input + output)
- ~7,800
- Tokens per model call
- ~2,700
- Model calls, all 9 runs
- 26
- Input tokens read from the prompt cache
- 88%
- Other hosts it tried to reach (the proxy refused them)
- models.dev
| Task (Claude Opus 5.5) | Passed | Median time |
|---|---|---|
| Create a file with given text | 3 / 3 | 8.1 s |
| Read a report and write a 2-sentence summary | 3 / 3 | 10.3 s |
| Count the data rows in a CSV with a shell command | 3 / 3 | 11.6 s |
Same three tasks, Claude Opus 5.5 driving each agent
| Measured | Goose | Hermes Agent | OpenCode |
|---|---|---|---|
| Tasks passed with Claude Opus 5.5 (of 9) | 9 | 9 | 9 |
| Median time: Create a file with given text (s) | 8.1 | 18.8 | 11 |
| Median time: Read a report and write a 2-sentence summary (s) | 10.3 | 24.8 | 16.3 |
| Median time: Count the data rows in a CSV with a shell command (s) | 11.6 | 22.3 | 10.9 |
| Model calls, all 9 runs | 26 | 24 | 25 |
| Tokens per task, input + output (approx.) | 7,800 | 47,100 | 29,200 |
| API cost, all 9 runs (US$, list price) | 0.14 | 0.27 | 0.19 |
| Outside hosts it tried to reach besides the API (blocked) | 1 | 0 | 2 |
| For reference: tasks passed with qwen2.5:3b on 3 CPU cores (of 9) | 5 | 0 | 3 |
All three ran on October 16, 2026, one after another, with Claude Opus 5.5 (claude-opus-5-5) through Anthropic's API and each agent's default settings: three runs per task, each in a fresh container, graded on the files left behind. The containers had no internet; a proxy let them reach api.anthropic.com and nothing else. Times include the round trips to Anthropic's API, so they show the agent and the API together, not this machine's speed. Tokens come from each agent's own output (Hermes Agent's model-call count is approximate); cost is those tokens at Anthropic's list price. The qwen2.5:3b row is from the earlier local-model runs (Goose and Hermes Agent on October 15, OpenCode on October 16).
What we noticed
- With qwen2.5:3b, every failure was a model mistake (reading a .txt file as an image, counting the CSV header row). Goose itself exited cleanly on all 9 runs. With Claude Opus 5.5 it passed all 9.
- It tried to reach models.dev on every run, with both models. We blocked it and Goose kept working.
Measured against Hermes Agent
| Measured | Goose | Hermes Agent |
|---|---|---|
| Tasks passed with qwen2.5:3b (of 9) | 5 | 0 |
| Fixed prompt sent with every model call (tokens) | 1,150 | 11,500 |
| Docker image size on disk (GB) | 0.71 | 4.17 |
| Outside hosts it tried to reach | 1 | 5 |
Same model (qwen2.5:3b on 3 CPU cores), same three tasks, three runs each, graded on the files left behind. Ollama capped the context at 32,768 tokens for both, below the 64K Hermes' docs ask for, and a 3B model is far below the model size Hermes asks for, so read the task score as 'which agent copes with a small local model', not overall quality.
Test machine: 8-vCPU AMD EPYC 9354P cloud VM, each test limited to 3 cores and 10 GB RAM, no GPU, no internet (isolated Docker network). Times in seconds are from this CPU-only machine; a desktop with a GPU will be much faster. The VM is shared, so text-generation speed is shown only as a ratio between tools run back to back, not as tokens per second. Docker image size is the on-disk size from docker image ls for every tool. Local AI tools we tested · How we test · raw data
Find your fit
Before you choose Goose
More options in AI Agent Apps. Compare pricing, platforms, and features side by side.
adris.tech
Free
Agencies and consultancies
Pros
- Passed 5 of 9 tasks with a 3B local model, the best of the three agents we tested.
- Lowest token use with Claude Opus 5.5: about 7,800 tokens per task and $0.14 for all 9 runs.
- A small fixed prompt (about 1,150 tokens), a tenth of Hermes Agent's.
- Desktop app, CLI and API, with MCP extensions and recipes you can reuse.
Cons
- With a 3B model it got only 1 of 3 report summaries and 1 of 3 CSV counts right, so plan on a bigger model for real work.
- It contacts models.dev on every run (it kept working when we blocked it).
- Running Goose locally doesn't keep your data local by itself; that depends on the model provider and extensions you connect.
About
- Platforms
- macOS, Linux, Windows
- API Available
- Yes
- Self-Hosted
- Yes
- Last Updated
- June 20, 2026
Best For
Who it's for
- Developers who want an open-source agent for file and shell tasks
What it does well
- Run file and shell tasks with the model you choose
- Save an MCP-powered workflow as a reusable recipe
Integrations
Frequently Asked Questions
- Is Goose free?
- Goose offers a free option. Hosting, model usage, or optional services may have separate costs; check the pricing details and official terms.
- Is Goose open source?
- Yes. Goose is open source.
- Does Goose have an API?
- Yes. Goose exposes a developer API. See the official documentation at https://goose-docs.ai for details.
- Can I self-host Goose?
- Yes. Goose supports self-hosting on your own infrastructure.
- When was Goose released?
- Goose was first released in 2025.
- What platforms does Goose support?
- Goose is available on: macOS, Linux, Windows.
People also compared
Curated lists that include this category
Goose is an open-source AI agent you run as a desktop app, command line or API. Connect a model and MCP extensions, and it can edit files, run commands and work with other services.
We ran Goose 1.54.0 on three small jobs: create a file, summarize a report, and count the rows in a CSV with a shell command, three runs each in fresh containers. With a small local model (qwen2.5:3b on 3 CPU cores) it passed 5 of 9, the best small-model score of the three agents we tested. Every failure was the model's mistake, such as counting the CSV header row, and Goose itself exited cleanly on every run. With Claude Opus 5.5 it passed all 9 and was the cheapest agent to run: about 7,800 tokens per task and $0.14 for all nine runs, against $0.19 for OpenCode and $0.27 for Hermes Agent. It was also the fastest on two of the three tasks. Recipes let you save a task's instructions and extension settings to reuse or share.
Of the three agents we tested, Goose did the most with a small local model and used the fewest tokens with Claude Opus 5.5. A strong pick for file and shell tasks; just check where each extension and model provider sends your data.
Contribute
- Work on Goose? Claim this listing.
- Embed a badge for your README or site.
Sign in to rate this tool or suggest an edit.