boardroom-map/startos/file-models/config.ts

import { FileHelper, z } from '@start9labs/start-sdk'

/**
 * Boardroom Map configuration, persisted to the `main` volume as config.json.
 *
 * Written by the StartOS actions (Configure Sparks / Models / Reviewers /
 * Review) and read by the Python orchestrator inside the container, which mounts
 * the same volume at /data and reads /data/config.json. Keep field names in sync
 * with orchestrator/bm_config.py (CONFIG_DEFAULTS).
 *
 * Boardroom Map is a CONTROL PLANE: a GPU-free orchestrator on StartOS that SSHes into
 * one or two DGX Sparks to serve local models and run a panel of sandboxed
 * "reviewer" containers over confidential documents you drop in. There is NO
 * frontier model and NO cloud key — everything stays on your hardware. The only
 * secrets are the Spark SSH key and an optional Hugging Face token (secrets.ts).
 */
export const configShape = z.object({
  // --- Spark connection (mirrors LLaMA-Factory / Nightshift) ---
  primarySparkHost: z.string().default(''),
  primarySparkUser: z.string().default('nvidia'),
  sshPort: z.number().int().positive().default(22),
  // Second Spark for extra model capacity. null = single node.
  secondarySparkHost: z.string().nullable().default(null),
  useBothSparks: z.boolean().default(false),
  // Address the head Spark's own preflight checks use to reach the local proxy.
  headInternalHost: z.string().default('127.0.0.1'),

  // --- Remote execution ---
  remoteWorkDir: z.string().default('/home/nvidia/boardroom-map'),

  // --- Images (built ON the Sparks; not packed into the s9pk) ---
  servingImage: z.string().default('boardroom-vllm:latest'),
  graderImage: z.string().default('boardroom-grader:latest'),

  // --- Serving (vLLM on the Sparks) ---
  gpuMemoryUtilization: z.string().default('0.85'),
  maxModelLen: z.number().int().positive().default(32768),
  // vLLM tool-call parser for native function-calling (the reviewer's read-file
  // tool loop relies on it). Must match the served model family — Qwen3 →
  // 'hermes'. Empty disables native tool-calling (reviewers fall back to a
  // JSON-action text protocol).
  toolCallParser: z.string().default('hermes'),
  // LiteLLM router exposing every model alias on one OpenAI-compatible endpoint.
  proxyPort: z.number().int().positive().default(4000),
  // How many distinct models may be co-resident on the HEAD Spark at once. The
  // job runner loads models in WAVES so it never exceeds this — letting you run a
  // panel across more models than fit in GPU memory simultaneously. 1 is safest.
  maxConcurrentModels: z.number().int().positive().default(1),

  // The MODEL CATALOG: the set of local models the service can serve. Each
  // reviewer (below) references one of these by `alias`. Mirrors
  // bm_config.py CONFIG_DEFAULTS["models"].
  models: z
    .array(
      z.object({
        alias: z.string(),
        hfModel: z.string(),
        // Which Spark serves this model. In `airgapped` network mode all models
        // must be on the head Spark (see networkMode).
        spark: z.enum(['primary', 'secondary']).default('primary'),
        port: z.number().int().positive().default(8001),
      }),
    )
    .default([
      { alias: 'reviewer-a', hfModel: 'Qwen/Qwen3-32B-FP8', spark: 'primary', port: 8001 },
    ]),

  // --- The review panel: one entry per reviewer ("number of reviews") ---
  // Each reviewer is a model + a persona (the lens it reads through) + an
  // optional temperature. Mirrors bm_config.py CONFIG_DEFAULTS["reviewers"].
  reviewers: z
    .array(
      z.object({
        name: z.string(),
        // Must match one of the model catalog aliases above.
        model: z.string(),
        persona: z.string().nullable().default(''),
        temperature: z.number().nullable().default(null),
      }),
    )
    .default([
      { name: 'reviewer-1', model: 'reviewer-a', persona: '', temperature: null },
    ]),

  // --- Review job settings ---
  // The rubric: what every reviewer should look for / produce. Layered above
  // each reviewer's persona.
  reviewInstructions: z.string().default(
    'Review the attached document(s). Produce a structured report: a 3-5 sentence ' +
      'summary, the key findings and insights, risks or red flags, open questions, ' +
      'and concrete recommendations. Cite the document and section for each point. ' +
      'Be honest about uncertainty; never invent facts not present in the documents.',
  ),
  // Confidentiality posture for the reviewer containers:
  //   'airgapped'      — reviewers join an --internal Docker network: they can
  //                      reach ONLY the on-Spark model proxy, with zero internet
  //                      egress. Models must be pre-pulled into the Spark's HF
  //                      cache (no live download). All models must be on the head
  //                      Spark. Strongest confidentiality.
  //   'local_services' — reviewers may also reach configured LAN services
  //                      (e.g. SearXNG) and the second Spark. NOTE: this network
  //                      has egress unless you firewall it — use only when you
  //                      accept that reviewers can reach the network.
  networkMode: z.enum(['airgapped', 'local_services']).default('airgapped'),
  // SearXNG JSON endpoint, used ONLY in local_services mode to give reviewers a
  // web_search tool. Empty = no web search.
  searxngUrl: z.string().default(''),

  // --- Synthesis (a local lead reviewer; no frontier model) ---
  synthesisEnabled: z.boolean().default(true),
  // Alias of the model that writes the consolidated report. Empty = first model.
  synthesisModel: z.string().default(''),
  // Optional persona/instructions for the lead reviewer. Empty = built-in default.
  synthesisPersona: z.string().default(''),

  // --- Document handling ---
  // After a job, wipe the extracted document text from the Sparks. Reports are
  // kept on the StartOS box regardless. Default true for confidentiality.
  wipeRemoteDocs: z.boolean().default(true),
  // Watch /data/inbox and auto-start a review when files land (debounced).
  // Default false: you trigger reviews explicitly with "Run Review".
  autoRunOnDrop: z.boolean().default(false),
  // Name of the per-job Docker network created on the head Spark.
  networkName: z.string().default('boardroom-net'),

  // --- Auth flags (the secret itself lives in secrets.ts) ---
  hfTokenSet: z.boolean().default(false),
})

export type Config = z.infer<typeof configShape>

export const configFile = FileHelper.json('./config.json', configShape)