13 orchestrators, one feature rubric
This is a weighted feature assessment from docs, repos and feature testing, not a measured performance leaderboard. Click a column to sort. "n/r" marks a criterion with no evidence: it is unranked, and the range shows the score if it were 0 or full marks. The assessed point score is the midpoint of that range.
Read the scores as a feature assessment. Orca was run end-to-end with stub agents; T3 Code was not run through an equivalent session, so Orca 93 vs T3 Code 90 is a difference in assessed features and evidence, not an experimentally proven gap.
- Doc coverage
- Share of rubric weight backed by documentation or other cited evidence.
- Hands-on validation
- None, Launched (installed and opened on Linux), or Stub end-to-end (six stub-agent sessions driven and verified).
- Measured results
- Which logged measurements exist for the tool in runs/.
Every criterion
| 1 | Orca | 5 | 5 | 4 | 5 | 5 | 4 | 5 | 5 | 4 | 93.0 | 93.0 to 93.0 | 100% | Stub end-to-end | Stub run x3, launch probe |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2 | T3 Code | 5 | 4 | 5 | 5 | 5 | 4 | 4 | 5 | 3 | 90.0 | 90.0 to 90.0 | 100% | Launched | Server probe |
| 3 | Synara | 5 | 4 | 4 | 3 | 5 | 3 | 3 | 5 | 4 | 79.0 | 79.0 to 79.0 | 100% | Launched | Launch probe |
| 4 | Emdash | 4 | 4 | 3 | 3 | 5 | 4 | 3 | 5 | n/r | 74.5 | 72.0 to 77.0 | 95% | Launched | Launch probe |
| 5 | Paseo | 4 | n/r | n/r | 5 | 5 | 4 | 4 | 5 | n/r | 74.5 | 57.0 to 92.0 | 65% | Launched | Launch probe |
| 6 | OpenChamber | 3 | 4 | n/r | 4 | 5 | 4 | 4 | 5 | n/r | 74.0 | 64.0 to 84.0 | 80% | Launched | Launch probe |
| 7 | Superset | 4 | 4 | 4 | 4 | 3 | 4 | 2 | 4 | n/r | 72.5 | 70.0 to 75.0 | 95% | Launched | Launch probe |
| 8 | Cloudroom | 4 | 4 | n/r | 5 | 5 | 3 | 2 | 5 | n/r | 72.0 | 62.0 to 82.0 | 80% | Launched | Launch probe |
| 9 | Vibe Kanban | 4 | 4 | 3 | 2 | 5 | 2 | 4 | 5 | 3 | 69.0 | 69.0 to 69.0 | 100% | Launched | Boot only |
| 10 | Claude Squad | 5 | 3 | 2 | 1 | 4 | 3 | 4 | 5 | 3 | 65.0 | 65.0 to 65.0 | 100% | Stub end-to-end | Stub run x3 |
| 11 | Conductor | 3 | 4 | 5 | 4 | 1 | n/r | 1 | 3 | 4 | 62.5 | 55.0 to 70.0 | 85% | None | None |
| 12 | Sculptor | 2 | 5 | 3 | 2 | 4 | 2 | n/r | 4 | 3 | 60.0 | 55.0 to 65.0 | 90% | None | None |
| 13 | Mux (now 'Xum') | n/r | 3 | n/r | n/r | 4 | 3 | 3 | 4 | n/r | 58.5 | 36.0 to 81.0 | 55% | Launched | Launch probe |
- A Agent breadth w15
- B Isolation and parallelism w15
- C Review, merge and PR flow w15
- D Remote, cloud and mobile reach w10
- E Openness and license w10
- F Maturity and momentum w15
- G Linux setup friction (measured) w10
- H Pricing and BYO model w5
- I Docs and transparency w5
Does the ranking survive other weightings?
| Weight profile | Top 5 (score) |
|---|---|
| Default (ours) | Orca (93.0), T3 Code (90.0), Synara (79.0), Emdash (74.5), Paseo (74.5) |
| Solo terminal hacker | Orca (94.5), T3 Code (88.1), Synara (79.4), OpenChamber (77.1), Emdash (76.3) |
| Team shipping PRs | T3 Code (90.4), Orca (90.4), Synara (77.4), Superset (74.3), Emdash (71.7) |
| Remote / phone-first | Orca (93.9), T3 Code (92.2), Paseo (78.7), Cloudroom (78.3), Synara (76.5) |
Solo terminal hacker: friction 20, openness 15, review 8, reach 3. Team shipping PRs: review 30, maturity 20. Remote / phone-first: reach 30, friction 5. Unlisted criteria keep the default weight.
9 more tools we checked
| Tool | Category | License | Why no score |
|---|---|---|---|
| Cursor (Agents Window) | First-party IDE | Proprietary | Not tested: needs account sign-in. |
| Codex (CLI + ChatGPT app) | First-party | Apache-2.0 (CLI); app proprietary | Not tested: no ChatGPT login on box. |
| Claude Code desktop | First-party | Proprietary | Not tested: no Claude login. |
| opencode | Agent CLI | MIT | Installed and used as the real-model agent in lane B. |
| Warp (Oz) | Terminal + cloud agents | AGPL-3.0 (client) | Not tested: needs account. |
| Amp | Agent + remote runtime | Proprietary (not verified) | Not tested: needs account. |
| Crystal -> Nimbalyst | Deprecated | MIT (Crystal) | Not tested: Crystal deprecated (last release v0.3.5, Feb 26 2026). |
| Terragon | Shut down | Apache-2.0 (as-is snapshot) | Not tested: service is gone; snapshot has no releases. |
| Omnara | Pivoted | Apache-2.0 | Not tested: it pivoted from mobile remote-control for coding CLIs to a managed-agent platform, a different category. |