ACP UIvsTessera
Generated from the two spec rows. Green marks the better value where a spec has a clear direction. Everything else is just different.
ACP UI
Jun Han · Agent harness
OSS
Panel
6.31 spec wins
- Reliability
- 5.8
- Usefulness
- 5.7
- Cost
- 8.3
- Longevity
- 5.2
“There is an Android build, so you can now approve a file write from the bus and regret it before your stop.”
Tessera
Horang Labs · Agent harness
OSS
Panel
6.32 spec wins
- Reliability
- 5.8
- Usefulness
- 6.5
- Cost
- 6.8
- Longevity
- 5.8
“A paired phone continues the same sessions, so the agent can now wait for your approval while you are asleep.”
Spec by spec
| Spec | ACP UI | Tessera |
|---|---|---|
| Architecture | ||
| Category | Agent harness | Agent harness |
| Runs | local | local |
| Platforms | macos, linux, windows, web | macos, linux, windows, web |
| Context window | not documented | not documented |
| Protocols | ||
| MCP client | No | No |
| MCP server | No | No |
| Capabilities | ||
| Runs terminal commands | YesAs the ACP client, ACP UI executes the terminal commands the agent requests and shows their output; the desktop build only, since the browser cannot spawn processes. | Yes |
| Multi-file edits | YesFile reads and writes are performed by ACP UI on the agent's behalf against the host workspace, and only on desktop. | YesFiles are changed by the Claude Code, Codex or OpenCode session Tessera runs in each worktree; Tessera adds its own file and diff editor on top. |
| Git operations | No | Yes |
| Browser control | No | No |
| Sandboxed execution | No | No |
| Multi-agent orchestration | No | YesA lead agent can create worktrees and launch, read and prompt parallel worker sessions through the /tessera-cli command. |
| Headless / CI mode | No | No |
| Models | ||
| Backbone | Claude Code, Codex, GitHub Copilot, Gemini CLI, Qwen Code, OpenCode, OpenClaw, Kiro CLI, any ACP-compatible agent | Claude Code, Codex, OpenCode |
| Bring your own model | Yes | Yes |
| Local models | No | No |
| Cost | ||
| Pricing model | byok | byok |
| Starts at | $0/mo | $0/mo |
| Free tier | Yes | Yes |
| Bring your own key | Yes | Yes |
| Openness | ||
| Open source | Yes | Yes |
| License | MIT | AGPL-3.0 |
| GitHub stars | 491 | 311 |
Which one would each critic pick
| Critic | ACP UI | Tessera | Pick |
|---|---|---|---|
| El Juez | — | — | not enough reviews |
| El Amigo | 6.5 | 6.8 | Tessera — Pick Tessera if you want agent work to move through visible stages; pick a single session if a board with five columns would only be a board you stopped updating. |
| El Crítico | 6.0 | 5.3 | ACP UI — It runs no agent of its own, so every capability listed belongs to somebody else's binary, and a lagging protocol implementation upstream turns this into an empty window. |
| El Profesor | 6.5 | 6.5 | no preference |
| La Inversora | 5.5 | 5.8 | Tessera — 299 stars around a control surface for three other companies' agents: the coordination is the contribution, and the vendors underneath are building their own. |
| La Jefa | 6.0 | 5.8 | ACP UI — Nothing per seat across sixty desks and nothing to govern: no console, no single sign-on, no audit record, and nothing that runs in a pipeline. |
| El Hacker | 7.0 | 7.5 | Tessera — Strong copyleft, which is the licence I want on something I depend on, and each backend takes a custom model identifier so I decide what is actually behind each session. |
Picks are derived from each critic's own scores. Humans vote on matchups on the duels page.