Choose Picasso.
You want the whole agent in one place: memory, tools, subagents, providers, goals, and trust controls.
picasso / compare / copilot-cliindexed
Compare
Picasso vs Copilot CLI
Plans and fleets, without the GitHub-only frame.
Choose PicassoYou want the whole agent in one place: memory, tools, subagents, providers, goals, and trust controls.
Choose Copilot CLIYou want its native surface and do not need Picasso's broader harness.
Run bothStart with one repository. Keep your current tool. Let the work decide.
CapabilityCopilot CLIPicassoPlan-first workflow.Yes.Yes.Fleet.Yes.Named and fork subagents.Persistent memory.No.Yes.
picasso mcp install githubpicasso auth provider add openai --subscriptionpicasso subagents listThe decision.
Copilot CLI belongs naturally in GitHub workflows. Picasso is for teams that want the same command-line discipline across providers, tools, memory, and longer work.
Choose Copilot CLI.
You want its native surface and do not need Picasso's broader harness.
Run both.
Start with one repository. Keep your current tool. Let the work decide.
Capability matrix.
| Capability | Copilot CLI | Picasso |
|---|---|---|
| Plan-first workflow. | Yes. | Yes. |
| Fleet. | Yes. | Named and fork subagents. |
| Persistent memory. | No. | Yes. |
| Provider choice. | GitHub models. | Anthropic, OpenAI, Google, xAI, Microsoft Foundry, Qwen, Xiaomi MiMo, Kimi, DeepSeek, Zhipu GLM, MiniMax, OpenRouter, Ollama, vLLM. |
| Goal mode. | No. | Yes. |
| MCP. | GitHub-centered. | Native client and server. |
A different center.
Picasso puts the loop at the center. Plan. Act. Observe. Reflect. Remember. The important state is visible, inspectable, and carried forward.
Before writes.
A plan shows the files, risks, checks, and open questions.
During work.
Tools, browser checks, MCP servers, and subagents stay inside the same loop.
After work.
Memory keeps the facts and decisions worth bringing back.
Where Picasso fits.
Beyond one platform.
Run the same harness on repositories, providers, and tools that are not GitHub-bound.
Memory that compounds.
Carry decisions and codebase facts across repeated sessions.
Goals with a finish line.
Budgets, stop conditions, and re-planning are built in.
Where Copilot CLI still fits.
GitHub-native policy.
Teams standardized on GitHub identity, repositories, and model routing may prefer Copilot CLI.
Repository-only flow.
If everything starts and ends in GitHub, the native integration is a natural fit.
Start here.
picasso mcp install githubpicasso auth provider add openai --subscriptionpicasso subagents listpicasso goal create "resolve the open issue queue with tests"Keep going.
Full matrix.
Compare model lock-in, memory depth, local execution, pricing path, and trust boundaries in one table.
Install.
Run Picasso against a task your current agent already handled and compare the plan, diff, checks, and replay.
Trust.
Read exactly what Picasso sees and what it never touches.