bo-next-batch

Name: bo-next-batch
Author: ossaidqadri

Generate next experimental suggestions from current run state.

by ossaidqadri(0) 0 installs

Free

Free account. Installing gives you the manifest plus copy-paste snippets.

See reviews

About

Imported from ossaidqadri/AI-Policy-Simulation (agentic-bo/.agents/skills/bo-next-batch/SKILL.md). Install upstream with npx skills add ossaidqadri/AI-Policy-Simulation --skill bo-next-batch. Copyright stays with the author.

BO Next Batch

Use this skill to get candidate experiments.

Command

uv run python -m bo_workflow.cli suggest --run-id <RUN_ID> --batch-size <N>

Accepts status initialized or running. No oracle needed — HEBO/BO/random work directly from the design space.

Return

list of suggestions with suggestion_id
feature values for each candidate
reminder to record outcomes later via bo-record-observation

Use it

Copy one of these into your project. Installing also returns the manifest and these snippets.

# after Install: the listing is in your workspace's routing pool - a plan picks it for its slot
curl -s -X POST https://api.opensmartroute.ai/api/v1/route -H 'Authorization: Bearer $OSR_API_KEY' -H 'Content-Type: application/json' -d '{"text": "...", "plan": true}'

Manifest

An Open Capability Manifest: the router reads it to know what this does, what it costs and when to pick it.

ossaidqadri-ai-policy-simulation-bo-next-batch.ocm.jsonjson

{
  "ocm": "1",
  "id": "ossaidqadri-ai-policy-simulation-bo-next-batch",
  "kind": "skill",
  "name": "bo-next-batch",
  "description": "Generate next experimental suggestions from current run state.",
  "publisher": "ossaidqadri",
  "version": "1.0.0",
  "capabilities": {
    "domains": [
      "general"
    ],
    "tags": [
      "skill-md",
      "github"
    ],
    "languages": [
      "en"
    ]
  },
  "quality_prior": 0.6,
  "examples": [
    "Generate next experimental suggestions from current run state."
  ],
  "primary": false,
  "metadata": {
    "source": {
      "provider": "github",
      "repository": "https://github.com/ossaidqadri/AI-Policy-Simulation",
      "path": "agentic-bo/.agents/skills/bo-next-batch/SKILL.md",
      "ref": "9dce43eaaea1c0c31c90690b1b7ca64b09691fe5",
      "url": "https://github.com/ossaidqadri/AI-Policy-Simulation/blob/9dce43eaaea1c0c31c90690b1b7ca64b09691fe5/agentic-bo/.agents/skills/bo-next-batch/SKILL.md",
      "key": "ossaidqadri/AI-Policy-Simulation/agentic-bo/.agents/skills/bo-next-batch/SKILL.md"
    }
  },
  "instructions": "# BO Next Batch\n\nUse this skill to get candidate experiments.\n\n## Command\n\n```bash\nuv run python -m bo_workflow.cli suggest --run-id <RUN_ID> --batch-size <N>\n```\n\nAccepts status `initialized` or `running`. No oracle needed — HEBO/BO/random work directly from the design space.\n\n## Return\n\n- list of suggestions with `suggestion_id`\n- feature values for each candidate\n- reminder to record outcomes later via `bo-record-observation`",
  "cost": {
    "context_tokens": 108
  }
}

Fetch it by URL: GET /api/v1/registry/ossaidqadri-ai-policy-simulation-bo-next-batch/manifest?version=1.0.0