@jmanuelcorral/openteam 0.25.0 → 0.26.0

package/README.es.md

@@ -41,7 +41,7 @@ estrecha un agente a `local` bajo el modo global `mixed`.
 ## Características
 
 - Routing por turno desde el hook `chat.message`, mutando `output.message.model` antes de que opencode persista el mensaje.
-- Registro de runtimes locales OpenAI-compatible: Ollama, LM Studio, Lemonade Server y Foundry Local.
+- Registro de runtimes locales OpenAI-compatible: Ollama, LM Studio, Lemonade Server, llama-swap y Foundry Local.
 - Perfiles de capacidad desde Models.dev, con cache y fallback curado cuando el catálogo no está disponible.
 - Selección automática **cheapest-capable** por coste, disponibilidad, soporte de tools y contexto dentro del dominio permitido.
 - Presupuestos por sesión, mes y tokens frontier por sesión.
@@ -62,6 +62,7 @@ estrecha un agente a `local` bajo el modo global `mixed`.
   - Ollama en `http://localhost:11434/v1`.
   - LM Studio en `http://localhost:1234/v1`.
   - Lemonade Server en `http://localhost:13305/api/v1`.
+  - llama-swap en `http://localhost:8080/v1`.
   - Foundry Local con puerto dinámico descubierto por CLI en Windows.
 
 ## Instalación
@@ -114,7 +115,7 @@ la política del operador en `.opencode/openteam.json`.
 ## Quickstart: `openteam setup`
 
 La forma más rápida de dejar un repositorio listo es `openteam setup`. Detecta
-Ollama, LM Studio, Lemonade Server y Foundry Local —todos opcionales—, permite
+Ollama, LM Studio, Lemonade Server, llama-swap y Foundry Local —todos opcionales—, permite
 elegir modelos locales y el **frontier baseline** (cheapest-capable primero), y
 genera la configuración y los artefactos de agentes/comandos:
 
@@ -143,6 +144,14 @@ lista vacía tampoco permite introducir un ID supuesto. openteam no carga ni
 descarga modelos para hacer funcionar una selección. Consulta la
 [política y los límites del servidor](docs/adr/lemonade-active-models.md).
 
+**llama-swap también exige modelos activos.** La integración inicial admite
+IDs locales canónicos con estado `ready` verificable y sin perfil de
+enrutamiento activo, en un servidor sin autenticación restringido por la red.
+Los límites declarados de contexto y salida se trasladan durante setup a
+opencode para compactar; el soporte de herramientas desconocido no se inventa.
+Consulta la [guía de llama-swap](docs/es/guide/llama-swap.md), incluidos los
+límites de la comprobación no atómica del cliente.
+
 El asistente:
 
 1. Sondea runtimes locales y primero pregunta cuáles habilitar. Elegir
@@ -232,7 +241,7 @@ Para instalar y probar openteam en otro equipo que ya tenga [opencode](https://o
    # La opción puntual ya ejecutó setup en el paso anterior.
    ```
 
-   El asistente detecta Ollama / LM Studio / Foundry Local, te deja elegir los modelos locales (locales o en otra máquina de la red) o **ninguno** para funcionar solo con frontier, y el frontier baseline, y escribe la configuración.
+   El asistente detecta Ollama / LM Studio / Lemonade Server / llama-swap / Foundry Local, te deja elegir los modelos locales (locales o en otra máquina de la red) o **ninguno** para funcionar solo con frontier, y el frontier baseline, y escribe la configuración.
 
 3. **Autentica el provider frontier** que hayas elegido:
 
@@ -348,9 +357,9 @@ ejecutas desde un subdirectorio verás "provider not configured" y debes hacer
 | `router.localDefault` | `ModelRef \| null` | `null` | Modelo local principal. No configurado por defecto — ningún modelo local se enruta hasta que configures uno con `openteam setup` o manualmente. |
 | `local.runtimes` | `LocalRuntime[]` | `[]` | Runtimes locales a sondear. Vacío por defecto — ningún runtime está configurado hasta que ejecutes `openteam setup` o añadas una entrada manualmente. |
 | `local.runtimes[].id` | `string` | requerido | Identidad de instancia, única dentro del array. Libre (`lmstudio-a`, `lemonade-lan`) para que dos máquinas del mismo tipo sean distintas; si se omite `kind`, debe ser en sí mismo un tipo de runtime (retrocompatibilidad). |
-| `local.runtimes[].kind` | `"ollama" \| "lmstudio" \| "foundry-local" \| "lemonade"` opcional | migra desde `id` | Tipo de runtime que selecciona el adaptador. Opcional: si falta, se usa `id` como tipo, así toda config previa sigue funcionando sin cambios. |
+| `local.runtimes[].kind` | `"ollama" \| "lmstudio" \| "foundry-local" \| "lemonade" \| "llama-swap"` opcional | migra desde `id` | Tipo de runtime que selecciona el adaptador. Opcional: si falta, se usa `id` como tipo, así toda config previa sigue funcionando sin cambios. |
 | `local.runtimes[].enabled` | `boolean` | `true` | Habilita o deshabilita ese runtime. |
-| `local.runtimes[].baseURL` | URL opcional | según runtime/config | Endpoint OpenAI-compatible (`/v1` para Ollama/LM Studio, `/api/v1` para Lemonade); Foundry Local puede omitirlo si usa discovery. |
+| `local.runtimes[].baseURL` | URL opcional | según runtime/config | Endpoint OpenAI-compatible (`/v1` para Ollama/LM Studio/llama-swap, `/api/v1` para Lemonade); Foundry Local puede omitirlo si usa discovery. |
 | `local.runtimes[].discovery` | `"cli" \| "sdk" \| "manual"` opcional | sin default por item | Discovery de Foundry Local; el registry usa CLI si está configurado y no es `manual`. |
 | `local.runtimes[].defaultModel` | `ModelRef` | requerido | Modelo por defecto de ese runtime. |
 | `local.runtimes[].maxConcurrency` | entero positivo opcional | `4` | Límite de generaciones concurrentes de esta instancia (su número real de slots). Se aplica por instancia. |
@@ -920,6 +929,20 @@ foundry service status
 
 openteam extrae el puerto de la salida del CLI y normaliza el endpoint a `http://localhost:<PORT>/v1`.
 
+### llama-swap
+
+Prepara el modelo en llama-swap antes de ejecutar setup. Estas consultas solo
+leen el catálogo, la disponibilidad y el perfil activo:
+
+```powershell
+Invoke-RestMethod -Uri "http://localhost:8080/v1/models" -Method Get
+Invoke-RestMethod -Uri "http://localhost:8080/running" -Method Get
+Invoke-RestMethod -Uri "http://localhost:8080/api/profiles" -Method Get
+```
+
+Consulta la [guía de llama-swap](docs/es/guide/llama-swap.md) para la selección
+de modelos listos, límites de contexto/salida y transportes soportados.
+
 ### Telemetría
 
 ```powershell
@@ -1011,7 +1034,7 @@ Consulta [docs/compatibility.md](docs/compatibility.md). Resumen verificado el `
 
 | Superficie | Versión / contrato |
 | --- | --- |
-| openteam | `0.25.0`; versión actual del paquete |
+| openteam | `0.26.0`; versión actual del paquete |
 | `@opencode-ai/plugin` | `1.18.19` |
 | `@opencode-ai/sdk` | `1.18.19` |
 | Hook de routing | `chat.message`; no `chat.params` para cambiar modelo |

package/README.md

@@ -47,7 +47,7 @@ or narrow an agent to `local` under global `mixed`, when frontier is prohibited.
 ## Features
 
 - Per-turn routing from the `chat.message` hook, mutating `output.message.model` before opencode persists the message.
-- Registration of OpenAI-compatible local runtimes: **Ollama**, **LM Studio**, **Lemonade Server** and **Foundry Local**.
+- Registration of OpenAI-compatible local runtimes: **Ollama**, **LM Studio**, **Lemonade Server**, **llama-swap** and **Foundry Local**.
 - Capability profiles from [Models.dev](https://models.dev), with cache and a curated fallback when the catalog is unavailable.
 - **Cheapest-capable** automatic selection by cost, availability, tool support, and context requirements inside the allowed domain.
 - Budgets per session, per month and per frontier tokens per session.
@@ -70,9 +70,10 @@ or narrow an agent to `local` under global `mixed`, when frontier is prohibited.
   - Ollama at `http://localhost:11434/v1`.
   - LM Studio at `http://localhost:1234/v1`.
   - Lemonade Server at `http://localhost:13305/api/v1`.
+  - llama-swap at `http://localhost:8080/v1`.
   - Foundry Local with a dynamic port discovered via CLI on Windows.
 
-> Local runtimes are **optional**. You can run openteam **frontier-only**, or point it at an Ollama / LM Studio / Lemonade Server running on **another machine on your LAN**.
+> Local runtimes are **optional**. You can run openteam **frontier-only**, or point it at an Ollama / LM Studio / Lemonade Server / llama-swap running on **another machine on your LAN**.
 
 ## Installation
 
@@ -125,7 +126,7 @@ operator policy in `.opencode/openteam.json`.
 ## Quickstart: `openteam setup`
 
 `openteam setup` detects the installed local runtimes (Ollama, LM Studio,
-Lemonade Server, Foundry Local — all optional), lets you pick local models and
+Lemonade Server, llama-swap, Foundry Local — all optional), lets you pick local models and
 the **frontier baseline** (sorted cheapest-capable first), and writes the
 project configuration and generated agent/command artifacts:
 
@@ -153,6 +154,13 @@ or unverified models are not selectable, and an empty active list does not
 allow a guessed model ID. openteam does not load or unload models to make a
 selection work. See the [policy and server-side limits](docs/adr/lemonade-active-models.md).
 
+**llama-swap also requires active models.** The initial integration accepts
+canonical local model IDs with verified `ready` state and no active routing
+profile, on an unauthenticated, network-restricted server. Declared context
+and output ceilings feed setup's opencode compaction limits; missing tool
+support is not guessed. See the [llama-swap guide](docs/guide/llama-swap.md)
+for configuration and the non-atomic server-side boundary.
+
 The wizard:
 
 1. Probes local runtimes, then first asks which ones to enable. Choosing
@@ -244,7 +252,7 @@ To install and try openteam on another machine that already has [opencode](https
    # The one-off option already ran setup in the previous step.
    ```
 
-   The wizard detects Ollama / LM Studio / Foundry Local, lets you pick the local models (local or on another machine on the LAN) or **none** to run frontier-only, and the frontier baseline, then writes the configuration.
+   The wizard detects Ollama / LM Studio / Lemonade Server / llama-swap / Foundry Local, lets you pick the local models (local or on another machine on the LAN) or **none** to run frontier-only, and the frontier baseline, then writes the configuration.
 
 3. **Authenticate the frontier provider** you chose:
 
@@ -355,9 +363,9 @@ configured" and should `cd` to the project root first.
 | `router.localDefault` | `ModelRef \| null` | `null` | Primary local model. Not set by default — no local model is routed until you configure one via `openteam setup` or by hand. |
 | `local.runtimes` | `LocalRuntime[]` | `[]` | Local runtimes to probe. Empty by default — no runtime is configured until you run `openteam setup` or add an entry by hand. |
 | `local.runtimes[].id` | string | required | Instance identity, unique within the array. Free-form (`lmstudio-a`, `lemonade-lan`) so two machines of the same kind are distinct; when `kind` is omitted it must itself be a runtime kind (back-compat). |
-| `local.runtimes[].kind` | `"ollama" \| "lmstudio" \| "foundry-local" \| "lemonade"` optional | migrates from `id` | Runtime kind selecting the adapter. Optional: when absent, `id` is treated as the kind, so pre-existing configs keep working unchanged. |
+| `local.runtimes[].kind` | `"ollama" \| "lmstudio" \| "foundry-local" \| "lemonade" \| "llama-swap"` optional | migrates from `id` | Runtime kind selecting the adapter. Optional: when absent, `id` is treated as the kind, so pre-existing configs keep working unchanged. |
 | `local.runtimes[].enabled` | `boolean` | `true` | Enable or disable that runtime. |
-| `local.runtimes[].baseURL` | optional URL | per runtime/config | OpenAI-compatible endpoint (`/v1` for Ollama/LM Studio, `/api/v1` for Lemonade); Foundry Local may omit it when using discovery. |
+| `local.runtimes[].baseURL` | optional URL | per runtime/config | OpenAI-compatible endpoint (`/v1` for Ollama/LM Studio/llama-swap, `/api/v1` for Lemonade); Foundry Local may omit it when using discovery. |
 | `local.runtimes[].discovery` | `"cli" \| "sdk" \| "manual"` optional | none per item | Foundry Local discovery; the registry uses CLI when configured and not `manual`. |
 | `local.runtimes[].defaultModel` | `ModelRef` | required | Default model for that runtime. |
 | `local.runtimes[].maxConcurrency` | positive integer, optional | `4` | Concurrent-generation cap for this instance (its real slot count). Applied per instance. |
@@ -873,6 +881,20 @@ foundry service status
 
 openteam extracts the port from the CLI output and normalizes the endpoint to `http://localhost:<PORT>/v1`.
 
+### llama-swap
+
+Prepare the intended model with llama-swap before setup. These calls only read
+the catalog, readiness and profile state:
+
+```powershell
+Invoke-RestMethod -Uri "http://localhost:8080/v1/models" -Method Get
+Invoke-RestMethod -Uri "http://localhost:8080/running" -Method Get
+Invoke-RestMethod -Uri "http://localhost:8080/api/profiles" -Method Get
+```
+
+See the [llama-swap guide](docs/guide/llama-swap.md) for ready-only selection,
+declared context/output limits and supported transport boundaries.
+
 ### Telemetry
 
 openteam writes a hash-only `CostRecord` JSONL log (never raw prompts). Tail the most recent entries:
@@ -961,7 +983,7 @@ See [docs/compatibility.md](docs/compatibility.md). Verified summary as of `2026
 
 | Surface | Version / contract |
 | --- | --- |
-| openteam | `0.25.0`; current package version |
+| openteam | `0.26.0`; current package version |
 | `@opencode-ai/plugin` | `1.18.19` |
 | `@opencode-ai/sdk` | `1.18.19` |
 | Routing hook | `chat.message`; not `chat.params` for model changes |

package/dist/certificates/graph-release-certificate.json

@@ -34,12 +34,12 @@
       "idempotent": true
     },
     "performance": {
-      "p95Millis": 0.4849879999999871,
+      "p95Millis": 0.3665430000000356,
       "budgetMillis": 50,
       "samples": 100
     }
   },
   "verdict": "pass",
   "failedGates": [],
-  "digest": "c3f5875e459a914431dd62d1d51666849fcc9746e37b47b2280b114bad36cb6d"
+  "digest": "b10ff6b124af850edf12082169fc980f87190f4c06c950e7324a95f32447aa8a"
 }

package/dist/certificates/graph-shadow-certificate.json

@@ -20,7 +20,7 @@
       "rawFindings": 0
     },
     "overhead": {
-      "p95Millis": 0.34576800000002095,
+      "p95Millis": 0.277688999999981,
       "budgetMillis": 25,
       "samples": 1000
     },
@@ -35,5 +35,5 @@
   },
   "verdict": "pass",
   "failedGates": [],
-  "digest": "5a4acec761ee74f67ee576dc55913b3afbd8f6524133d3075022928daa6e8adf"
+  "digest": "0a32756f8be276eafc5ed1577804aa71f52bdae689a425c2d6566e8d45d0d0f1"
 }

package/dist/cli/setupAdapters.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"setupAdapters.d.ts","sourceRoot":"","sources":["../../src/cli/setupAdapters.ts"],"names":[],"mappings":"AAeA,OAAO,KAAK,EAAE,sBAAsB,EAAE,MAAM,uBAAuB,CAAC;AACpE,OAAO,KAAK,EACV,eAAe,EAEf,QAAQ,EAET,MAAM,mBAAmB,CAAC;AAE3B,OAAO,EACL,KAAK,QAAQ,EAEd,MAAM,wBAAwB,CAAC;AAIhC,OAAO,KAAK,EAAE,gBAAgB,EAAwB,MAAM,gBAAgB,CAAC;AAyB7E,qDAAqD;AACrD,eAAO,MAAM,aAAa,EAAE,QA8E3B,CAAC;AA6BF;;;GAGG;AACH,wBAAgB,0BAA0B,IAAI,MAAM,OAAO,CACzD,sBAAsB,EAAE,CACzB,CAcA;AACD,wBAAgB,YAAY,CAAC,IAAI,EAAE,QAAQ,GAAG,MAAM,OAAO,CAAC,eAAe,EAAE,CAAC,CAiB7E;AAqBD;;;GAGG;AACH,wBAAgB,kBAAkB,CAChC,IAAI,EAAE,QAAQ,GACb,CAAC,IAAI,EAAE,gBAAgB,EAAE,OAAO,EAAE,MAAM,KAAK,OAAO,CAAC,eAAe,CAAC,CAmBvE"}
+{"version":3,"file":"setupAdapters.d.ts","sourceRoot":"","sources":["../../src/cli/setupAdapters.ts"],"names":[],"mappings":"AAeA,OAAO,KAAK,EAAE,sBAAsB,EAAE,MAAM,uBAAuB,CAAC;AACpE,OAAO,KAAK,EACV,eAAe,EAEf,QAAQ,EAET,MAAM,mBAAmB,CAAC;AAE3B,OAAO,EACL,KAAK,QAAQ,EAEd,MAAM,wBAAwB,CAAC;AAKhC,OAAO,KAAK,EAAE,gBAAgB,EAAwB,MAAM,gBAAgB,CAAC;AAyB7E,qDAAqD;AACrD,eAAO,MAAM,aAAa,EAAE,QA8E3B,CAAC;AAmCF;;;GAGG;AACH,wBAAgB,0BAA0B,IAAI,MAAM,OAAO,CACzD,sBAAsB,EAAE,CACzB,CAcA;AACD,wBAAgB,YAAY,CAAC,IAAI,EAAE,QAAQ,GAAG,MAAM,OAAO,CAAC,eAAe,EAAE,CAAC,CAiB7E;AAqBD;;;GAGG;AACH,wBAAgB,kBAAkB,CAChC,IAAI,EAAE,QAAQ,GACb,CAAC,IAAI,EAAE,gBAAgB,EAAE,OAAO,EAAE,MAAM,KAAK,OAAO,CAAC,eAAe,CAAC,CAmBvE"}