Answers from the documentation; checking which model is on duty
Hi - I answer from the OpenSmartRoute documentation: routing, the API, plans and quotas, self-hosting. Ask away, or open a support ticket if you need a person.
Grounded in the docs - follow a source before acting on it.
kernel - Agent - OpenSmartRoute
Agentv1.0.0
kernel
Lead. Plans Kernel sprint work, splits issues that are too big, and keeps Linear honest. Use for planning, sequencing and cross-lane questions.
Claude Code subagent imported from cjjutba/kernel (.claude/agents/kernel.md). Copyright stays with the author.
You are Kernel, leading the work on Kernel itself. You plan, you don't write feature code.
Read AGENTS.md, docs/ROADMAP.md and the Linear issue before planning.
Break work so each piece fits one PR and stays inside one lane's files.
When an issue is too big, propose sub-issues in Linear with screens and acceptance criteria in the same shape as the parent.
Watch for lanes about to edit the same shared file and flag it before it happens.
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 - nothing else to configurecurl-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}'# or pin it on the OpenAI-compatible endpoint: {"model": "cjjutba-kernel-kernel-subagent", ...}
Installed into a catalogue, chosen by a router
Install kernel and it becomes one more candidate the router can pick - when it fits.
A listing is a routing target with a manifest: what it does, which domains it covers, what it costs and who publishes it. Once installed it sits beside your own models and tools, is scored like any other candidate for each request, and shows up in the trace when it wins. Ratings come from workspaces that installed it, one per account.
When a provider is down or over quota the request moves on; the trace says it did.
Manifest
An Open Capability Manifest: the router reads it to know what this does, what it costs and when to pick it.
cjjutba-kernel-kernel-subagent.ocm.jsonjson
{
"ocm": "1",
"id": "cjjutba-kernel-kernel-subagent",
"kind": "agent",
"name": "kernel",
"description": "Lead. Plans Kernel sprint work, splits issues that are too big, and keeps Linear honest. Use for planning, sequencing and cross-lane questions.",
"publisher": "cjjutba",
"version": "1.0.0",
"capabilities": {
"domains": [
"coding"
],
"tags": [
"agent-md",
"github-claude-agents"
],
"languages": [
"en"
]
},
"quality_prior": 0.6,
"examples": [
"Lead. Plans Kernel sprint work, splits issues that are too big, and keeps Linear honest. Use for planning, sequencing and cross-lane questions."
],
"primary": false,
"metadata": {
"source": {
"provider": "github-claude-agents",
"repository": "https://github.com/cjjutba/kernel",
"path": ".claude/agents/kernel.md",
"ref": "1ca6c3d16c762b6d7718a19d5e812f5b19d26cbf",
"url": "https://github.com/cjjutba/kernel/blob/1ca6c3d16c762b6d7718a19d5e812f5b19d26cbf/.claude/agents/kernel.md",
"key": "cjjutba/kernel/.claude/agents/kernel.md"
},
"tools": [
"Read",
"Grep",
"Glob",
"Bash",
"mcp__linear",
"mcp__claude_ai_Linear",
"mcp__pencil"
]
},
"instructions": "You are Kernel, leading the work on Kernel itself. You plan, you don't write feature code.\n- Read AGENTS.md, docs/ROADMAP.md and the Linear issue before planning.\n- Break work so each piece fits one PR and stays inside one lane's files.\n- When an issue is too big, propose sub-issues in Linear with screens and acceptance criteria in the same shape as the parent.\n- Watch for lanes about to edit the same shared file and flag it before it happens.",
"cost": {
"context_tokens": 111
}
}
Fetch it by URL: GET /api/v1/registry/cjjutba-kernel-kernel-subagent/manifest?version=1.0.0
Reviews
Star ratings from people who tried it. One review per account; edit yours any time.
No reviews yet. Install it, try it, and be the first to rate it.
