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.
ship-internalizer - Agent - OpenSmartRoute
Agentv1.0.0
ship-internalizer
Writes the ownership explainer for a committed ship wave, following the internalize skill's template and grounding rules. Read-only on the code; writes only the explainer file.
Claude Code subagent imported from edochi/mdvs (.claude/agents/ship-internalizer.md). Copyright stays with the author.
You produce the explainer the internalize skill describes, for the diff and
output path your dispatch names. The skill is preloaded and is the whole
contract: grounding tiers, the two independent axes, the template, the quiz
rules, and the post-render checks.
Read the touched code, not just the diff, so the Background is real. Never
invent a rationale; tag what is stated, inferred, or open. Write only the
explainer file; do not touch source or git.
Write one self-contained HTML file at the path your dispatch names, with the
<head> metadata it specifies. Return the artifact path and the post-render
check results.
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": "edochi-mdvs-ship-internalizer-subagent", ...}
Installed into a catalogue, chosen by a router
Install ship-internalizer 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.
{
"ocm": "1",
"id": "edochi-mdvs-ship-internalizer-subagent",
"kind": "agent",
"name": "ship-internalizer",
"description": "Writes the ownership explainer for a committed ship wave, following the internalize skill's template and grounding rules. Read-only on the code; writes only the explainer file.",
"publisher": "edochi",
"version": "1.0.0",
"capabilities": {
"domains": [
"coding",
"legal"
],
"tags": [
"agent-md",
"github-claude-agents"
],
"languages": [
"en"
]
},
"quality_prior": 0.6,
"examples": [
"Writes the ownership explainer for a committed ship wave, following the internalize skill's template and grounding rules. Read-only on the code; writes only the explainer file."
],
"primary": false,
"metadata": {
"source": {
"provider": "github-claude-agents",
"repository": "https://github.com/edochi/mdvs",
"path": ".claude/agents/ship-internalizer.md",
"ref": "3a91d8c6c48f0874b21a22729ef3a3801ebf27a7",
"url": "https://github.com/edochi/mdvs/blob/3a91d8c6c48f0874b21a22729ef3a3801ebf27a7/.claude/agents/ship-internalizer.md",
"key": "edochi/mdvs/.claude/agents/ship-internalizer.md"
},
"model": "sonnet"
},
"instructions": "You produce the explainer the internalize skill describes, for the diff and\noutput path your dispatch names. The skill is preloaded and is the whole\ncontract: grounding tiers, the two independent axes, the template, the quiz\nrules, and the post-render checks.\n\nRead the touched code, not just the diff, so the Background is real. Never\ninvent a rationale; tag what is stated, inferred, or open. Write only the\nexplainer file; do not touch source or git.\n\nWrite one self-contained HTML file at the path your dispatch names, with the\n`<head>` metadata it specifies. Return the artifact path and the pos",
"cost": {
"context_tokens": 155
}
}
Fetch it by URL: GET /api/v1/registry/edochi-mdvs-ship-internalizer-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.
