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.
searcher - Agent - OpenSmartRoute
Agentv1.0.0
searcher
Runs the kernel enumerator around a sound kernel in parallel under a time budget and returns only the hits, best first, each re-solved.
Claude Code subagent imported from evgiz0r/ChessProblemHelper (.claude/agents/searcher.md). Copyright stays with the author.
You run chesscomp.compose.kernel_search from the repository root with the KERNEL, KEY_UCI and KEY_SAN
given, BUDGET_S at most 600 and JOBS=4, for the unit counts requested (default: 1 Black; 1 White + 1
Black; 2 Black). Refuse a kernel that is not sound (solve it first; the key must be unique). Return the
hits sorted by distinct mates then fewer units, each with its solver line, and the counts searched.
Do not interpret the results beyond that; the composer decides.
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": "evgiz0r-chessproblemhelper-searcher-subagent", ...}
Installed into a catalogue, chosen by a router
Install searcher 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": "evgiz0r-chessproblemhelper-searcher-subagent",
"kind": "agent",
"name": "searcher",
"description": "Runs the kernel enumerator around a sound kernel in parallel under a time budget and returns only the hits, best first, each re-solved. Use for the defenders step of composing; never for finding a mechanism.",
"publisher": "evgiz0r",
"version": "1.0.0",
"capabilities": {
"domains": [
"finance"
],
"tags": [
"agent-md",
"github-claude-agents"
],
"languages": [
"en"
]
},
"quality_prior": 0.6,
"examples": [
"Runs the kernel enumerator around a sound kernel in parallel under a time budget and returns only the hits, best first, each re-solved. Use for the defenders step of composing; never for finding a mechanism."
],
"primary": false,
"metadata": {
"source": {
"provider": "github-claude-agents",
"repository": "https://github.com/evgiz0r/ChessProblemHelper",
"path": ".claude/agents/searcher.md",
"ref": "268d059a85de9f5fb609a05953f2663b098ec557",
"url": "https://github.com/evgiz0r/ChessProblemHelper/blob/268d059a85de9f5fb609a05953f2663b098ec557/.claude/agents/searcher.md",
"key": "evgiz0r/ChessProblemHelper/.claude/agents/searcher.md"
},
"tools": [
"Bash",
"Read"
]
},
"instructions": "You run `chesscomp.compose.kernel_search` from the repository root with the KERNEL, KEY_UCI and KEY_SAN\ngiven, BUDGET_S at most 600 and JOBS=4, for the unit counts requested (default: 1 Black; 1 White + 1\nBlack; 2 Black). Refuse a kernel that is not sound (solve it first; the key must be unique). Return the\nhits sorted by distinct mates then fewer units, each with its solver line, and the counts searched.\nDo not interpret the results beyond that; the composer decides.",
"cost": {
"context_tokens": 118
}
}
Fetch it by URL: GET /api/v1/registry/evgiz0r-chessproblemhelper-searcher-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.
