Appearance
Orca agents and OrcaCompass support
Orca's observed agent support, and what OrcaCompass does with each agent.
The Orca launch and trust behavior below was inspected on 2026-11-03 with Orca IDE 1.4.246. It is a compatibility reference, not a guarantee for newer Orca versions. The OrcaCompass adapter descriptions reflect this checkout.
Every agent id below can be named in orca orchestration worker-start --agent <id>. The list comes from Orca's built-in agent table (resources/app.asar.unpacked/out/main/chunks/tui-agent-config-*.js in the Orca install).
Columns:
--modelat launch: from Orca's per-agent catalogs (resources/app.asar.unpacked/out/shared/agent-session/options/agent-session-option-catalog-*.js). ✅ Orca passes--model(and--effortwhere listed). ❌ Orca has a catalog but refuses--modelat launch (Agent <id> does not support launch-time model selection). — no catalog. With ❌ or —, the agent runs the model from its own config.- Auto-trust: Orca answers the agent's "trust this folder" prompt before launch (
preflightTrust), or the launch command carries a trust flag. Without it, a worker in a folder the agent has not trusted can stop at that prompt; Orca then fails the start withAgent startup blocked: agent-trust-workspace. - OrcaCompass: what OrcaCompass does with the agent today.
| Agent id | Command | --model at launch | Auto-trust | OrcaCompass |
|---|---|---|---|---|
claude | claude | ✅ | ✅ | Supported: routed |
claude-agent-teams | orca claude-teams (not on Windows or WSL) | not checked | ✅ | Not supported |
codex | codex | ✅ | ✅ | Supported: routed |
cursor | cursor-agent | ✅ (effort is part of the model id) | ✅ | Supported: routed; also carries xAI, Google, Meta and other models through Cursor |
antigravity | agy | ✅ | ✅ | Supported: routed |
muse | muse --trust-workspace | ✅ | ✅ (flag) | Orca accepts any --model; neither Muse nor Orca lists Muse models, so OrcaCompass routes a built-in list |
grok | grok | ✅ (since 1.4.246; passed as -m / --reasoning-effort) | ❌ | Supported: routed. OrcaCompass reads Grok's models (~/.grok/models_cache.json, then grok models), tiers them and knows their effort limits. Orca does not auto-trust folders for Grok, so a worker in a folder Grok has not trusted stops at its trust prompt. xAI models also route through Cursor |
hermes | hermes --tui | — | — | Not supported: Orca passes no model to Hermes |
gemini | gemini | ❌ | — | Not an agent here: Gemini models route through Antigravity or Cursor |
codebuddy | codebuddy / cbc | ✅ | — | Not supported yet (Orca side allows it) |
omp | omp | ✅ (no effort) | — | Not supported yet (Orca side allows it) |
opencode | opencode | ✅ (existing worktree only, no effort) | — | Supported: routed as a last resort by default (lastResortAgents). Models from a live opencode models call: API-key providers such as OpenRouter (pay per token), OpenCode's own models, and local servers such as Ollama. No effort is passed. Its routes need --worktree current or another existing worktree, because Orca refuses --model for OpenCode in a new one |
qoder | qodercli / qoder | — | ✅ | Not supported |
qoder-cn | qoderclicn / qodercn | — | ✅ | Not supported |
kimi | kimi / kimi-code | — | ✅ | Not supported |
copilot | copilot | — | ✅ | Not supported |
command-code | command-code --trust | — | ✅ (flag) | Not supported |
opencode2 | opencode2 --standalone | — | — | Not supported |
openclaude | openclaude | — | — | Not supported |
autohand | autohand | — | — | Not supported |
ante | ante | — | — | Not supported |
trae | traecli | — | — | Not supported |
mimo-code | mimo | — | — | Not supported |
pi | pi | — | — | Not supported |
prime-agent | prime-agent | — | — | Not supported |
aider | aider | — | — | Not supported |
goose | goose | — | — | Not supported |
amp | amp | — | — | Not supported |
kilo | kilo | — | — | Not supported |
kiro | kiro-cli chat --tui | — | — | Not supported |
crush | crush | — | — | Not supported |
aug | auggie | — | — | Not supported |
cline | cline | — | — | Not supported |
freebuff | freebuff | — | — | Not supported |
codebuff | codebuff | — | — | Not supported |
continue | cn | — | — | Not supported |
droid | droid | — | — | Not supported |
mistral-vibe | vibe / mistral-vibe | — | — | Not supported |
qwen-code | qwen | — | — | Not supported |
rovo | acli rovodev run | — | — | Not supported |
openclaw | openclaw | — | — | Not supported |
dsh | dsh-tui . | — | — | Not supported |
zcode | zcode | — | — | Not supported |
devin | devin | — | — | Not supported |
jcode | jcode | — | — | Not supported |
How OrcaCompass handles each agent
Source: src/routing/environment.ts and src/placements/agents/*.ts, src/routing/router.ts, src/routing/rejection.ts, src/verify.ts.
Shared steps (every agent)
- Find the CLI with
find(<binary>, extra dirs). Not found →installed: false, never routed. - Read the model list when the placement table is built (every
placementsTtlHours, onrefresh, or when the table is stale), never on a route that reuses a fresh table. For Claude Code and Codex, Orca's model catalog comes first (src/placements/agents/orca-catalog.ts): the newest probe inagent-model-catalog.dbthat found models for the agent, opened read-only once per build, then theagent-model-catalog.jsonolder Orca versions wrote. Orca saves no other agent's list there. Live model-list commands run throughlistModels(src/placements/agents/list-cache.ts): a good listing is saved to the OrcaCompass cache; if the command later fails, the saved copy is used with its age ("using the list from 3h ago"). A sign-in prompt in the output marks the agentsignedOutand never falls back. - Give each model its profile: coding, agentic and reasoning scores from the evidence snapshot, and a cost tier from your
tierOverrides, the snapshot price, the family or lineup hint, then the evidence class (src/placements/profiles.ts), stored inplacements.json(src/placements/build.ts). - Check usage with
orca account list --json(src/routing/orca-usage.ts): each model draws on named usage pools; a missing number is skipped, not read as 0 %. - Skip the agent when it is turned off in Orca's Settings → Agents (
disabledTuiAgents, read-only from Orca's profile state,src/routing/orca-settings.ts), when Orca reports itunavailable, or when it is signed out. - Skip models the agent refused at an earlier launch (
orcacompass verifysaves them forrejectionTtlHours). - Build
launchArgs(resolveinsrc/routing/router.ts):flagmodels:--agent <id> --model <id> [--effort <e>], effort snapped to the nearest one the model offers (applied: flag, orprovider-defaultwhen none).variantmodels (effort inside the id):--agent <id> --model <id-for-effort>, no--effort(applied: model-id). Explicit--effortrequests are never snapped: models that cannot apply that exact level are excluded.
- After launch,
orcacompass verify --dispatch <id>reads the worker's output and matches the agent's own refusal text (below).
Per agent
| Claude | Codex | Cursor | Antigravity | Grok | OpenCode | Muse | |
|---|---|---|---|---|---|---|---|
| Adapter | claude.ts | codex.ts | cursor.ts | antigravity.ts | grok.ts | opencode.ts | muse.ts |
| Finds | claude on PATH | codex on PATH | cursor-agent on PATH or %LOCALAPPDATA%\cursor-agent | agy on PATH, %LOCALAPPDATA%\agy\bin or ~/.local/bin | grok on PATH, $GROK_HOME/bin or ~/.grok/bin | opencode on PATH or ~/.opencode/bin | muse on PATH |
| Model list | Orca's agent-model-catalog.db, then Claude Code's cached account catalog, then Anthropic families in the evidence snapshot | Orca's agent-model-catalog.db, then $CODEX_HOME/models_cache.json (visibility: list), then the evidence snapshot's Codex ladder families without -latest aliases | cursor-agent models, live | agy models, live (id<TAB>label rows) | ~/.grok/models_cache.json (Grok's own, hidden models dropped); else grok models, live | opencode models, live (provider/model rows; OpenRouter ~ aliases skipped) | Built-in: muse-wonder-1.0, muse-spark-1.1, muse-ember-1.0, muse-spark-1.3 (MUSE_FALLBACK_MODELS) |
| Grouping | — | — | Variants grouped by base id: effort, -thinking and -fast are stripped; cursor- prefix and old claude-4.6-sonnet ids normalized | -low/-medium/-high ids grouped; -thinking dropped to match the evidence snapshot | — | A model listed by several providers kept once, from the cheapest (local, then free, then paid) | — |
| Effort | Orca's catalog levels + default, else the account catalog; on the evidence snapshot fallback, assumed low to max | Orca's catalog levels + default, else the Codex cache's; on the evidence snapshot fallback, assumed low to max | Inside the model id (…-high); the plain id stands for medium; fast-only levels dropped | Gemini: inside the id. Others: Orca's low/medium/high via --effort | Grok cache levels + default (none for multi-agent models); from grok models: low to xhigh, capped as in Orca's seed (4.5 up to high) | None (Orca passes no effort to OpenCode) | — |
launchArgs | --model --effort | --model --effort | --model <variant id> | --model <variant id> (Gemini) or --model --effort | --model --effort | --model <provider/model> | — |
| Usage pools | session, weekly, plus fableWeekly for Fable | session, weekly | monthly + Cursor Models or Other Models bucket (which models draw on which is inferred) | Per quota group (Gemini Models, Claude and GPT models): 5-hour and long; else Orca's session/weekly | weekly, monthly | None: pay per use or local; the route says how it is paid | — |
Refusal verify matches | There's an issue with the selected model (idles) | model is not supported when using Codex… (idles) | Cannot use this model (exits) | invalid model selection / not recognized as a known model (exits) | Couldn't set model … unknown model id (exits) | Model unavailable | — |
| Special rule | — | — | Cursor auto placed only in Light and Daily | — | — | Last resort by default (lastResortAgents); routes note that Orca needs an existing worktree | — |
Where OrcaCompass and Orca disagree today
- Grok: Orca 1.4.245 refused
--modelfor Grok workers; 1.4.246 accepts--agent grok --model <id> --effort <e>(stablyai/orca#36480) and still refuses effort levels above a model's ceiling (Grok 4.5 abovehigh), which matches the ceilings ingrok.ts. End-to-end validation remains incomplete. - Effort lists for Claude and Codex come from their local catalogs, Orca's catalog, or the the evidence snapshot fallback. Orca checks effort against its own catalog for models it knows (for example
gpt-5.5up toxhigh) and refuses others. Whether the two lists always agree has not been checked. - OpenCode: the worktree rule is only a note in
reasons;launchArgsdo not add--worktree, so the caller must pick--worktree currentor an existing one. - Folder trust: OrcaCompass does not check it. Orca handles it for Claude, Codex, Cursor and Antigravity (and Muse through its launch flag). Grok can stop at its trust prompt; OpenCode has no auto-trust in Orca, but whether it asks has not been checked.