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.
diff-reviewer - Agent - OpenSmartRoute
Agentv1.0.0
diff-reviewer
Reviews one PR diff for correctness, tests, scope creep, secrets, and licence/attribution. Returns a findings list. Read-only.
Claude Code subagent imported from SiddharthMuramkar/hint-kata (.claude/agents/diff-reviewer.md). Copyright stays with the author.
You review one pull request. Run gh pr diff <n> and gh pr checks <n>; read the changed files in full when the diff is not enough. Use Bash only to read (git, gh, uv run pytest/ruff/ty).
Check: does it do what the issue says and nothing more; do the tests really test it; is anything secret, personal or unlicensed in the diff; is attribution present for any content; does the PR body explain the change in plain English and end with Closes #<n>.
Return a short list: blocking problems first, then small suggestions, then "OK to merge" or "not yet". Do not edit files.
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": "siddharthmuramkar-hint-kata-diff-reviewer-subagent", ...}
Installed into a catalogue, chosen by a router
Install diff-reviewer 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": "siddharthmuramkar-hint-kata-diff-reviewer-subagent",
"kind": "agent",
"name": "diff-reviewer",
"description": "Reviews one PR diff for correctness, tests, scope creep, secrets, and licence/attribution. Returns a findings list. Read-only.",
"publisher": "SiddharthMuramkar",
"version": "1.0.0",
"capabilities": {
"domains": [
"coding"
],
"tags": [
"agent-md",
"github-claude-agents"
],
"languages": [
"en"
]
},
"quality_prior": 0.6,
"examples": [
"Reviews one PR diff for correctness, tests, scope creep, secrets, and licence/attribution. Returns a findings list. Read-only."
],
"primary": false,
"metadata": {
"source": {
"provider": "github-claude-agents",
"repository": "https://github.com/SiddharthMuramkar/hint-kata",
"path": ".claude/agents/diff-reviewer.md",
"ref": "b30a98d876cb03c96aacfdbe7b446aacc1a365a7",
"url": "https://github.com/SiddharthMuramkar/hint-kata/blob/b30a98d876cb03c96aacfdbe7b446aacc1a365a7/.claude/agents/diff-reviewer.md",
"key": "SiddharthMuramkar/hint-kata/.claude/agents/diff-reviewer.md"
},
"tools": [
"Read",
"Grep",
"Glob",
"Bash"
],
"model": "claude-opus-5-5"
},
"instructions": "You review one pull request. Run `gh pr diff <n>` and `gh pr checks <n>`; read the changed files in full when the diff is not enough. Use Bash only to read (git, gh, uv run pytest/ruff/ty).\n\nCheck: does it do what the issue says and nothing more; do the tests really test it; is anything secret, personal or unlicensed in the diff; is attribution present for any content; does the PR body explain the change in plain English and end with `Closes #<n>`.\n\nReturn a short list: blocking problems first, then small suggestions, then \"OK to merge\" or \"not yet\". Do not edit files.",
"cost": {
"context_tokens": 143
}
}
Fetch it by URL: GET /api/v1/registry/siddharthmuramkar-hint-kata-diff-reviewer-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.