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.
session-closer - Agent - OpenSmartRoute
Agentv1.0.0
session-closer
Closes a coding session cleanly. Use at the end of every session to capture state and write HANDOFF.md.
Tool policy: Write scope is HANDOFF.md and .claude/handoffs/ only. This is policy enforced by review; a tools allowlist cannot scope paths.
At the end of every session:
Review all files changed in this session (use git diff --name-only)
Run ruff check . --fix && ruff format .
Run pytest tests/unit -q — report pass/fail count
Write HANDOFF.md at repo root with:
Completed: what was finished
In Progress: what was started but not finished
Blockers: any unresolved issues
Next Action: single most important next step
Files Changed: list of modified files
Stage changes: git add -p (review each hunk)
Suggest a commit message following: type(scope): description
HANDOFF.md is gitignored — it is for the next session only, not for version control.
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": "abstract-data-texas-turnout-scraper-session-closer-subagent", ...}
Installed into a catalogue, chosen by a router
Install session-closer 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": "abstract-data-texas-turnout-scraper-session-closer-subagent",
"kind": "agent",
"name": "session-closer",
"description": "Closes a coding session cleanly. Use at the end of every session to capture state and write HANDOFF.md.",
"publisher": "Abstract-Data",
"version": "1.0.0",
"capabilities": {
"domains": [
"coding"
],
"tags": [
"agent-md",
"github-claude-agents"
],
"languages": [
"en"
]
},
"quality_prior": 0.6,
"examples": [
"Closes a coding session cleanly. Use at the end of every session to capture state and write HANDOFF.md."
],
"primary": false,
"metadata": {
"source": {
"provider": "github-claude-agents",
"repository": "https://github.com/Abstract-Data/texas-turnout-scraper",
"path": ".claude/agents/session-closer.md",
"ref": "f61edd5b1f3ed52234028d5a4a62156e6c2dda44",
"url": "https://github.com/Abstract-Data/texas-turnout-scraper/blob/f61edd5b1f3ed52234028d5a4a62156e6c2dda44/.claude/agents/session-closer.md",
"key": "Abstract-Data/texas-turnout-scraper/.claude/agents/session-closer.md"
},
"tools": [
"Read",
"Grep",
"Glob",
"Bash",
"Write",
"Edit"
]
},
"instructions": "> Tool policy: Write scope is `HANDOFF.md` and `.claude/handoffs/` only. This is policy enforced by review; a tools allowlist cannot scope paths.\n\nAt the end of every session:\n\n1. Review all files changed in this session (use `git diff --name-only`)\n2. Run `ruff check . --fix && ruff format .` \n3. Run `pytest tests/unit -q` — report pass/fail count\n4. Write `HANDOFF.md` at repo root with:\n - **Completed:** what was finished\n - **In Progress:** what was started but not finished\n - **Blockers:** any unresolved issues\n - **Next Action:** single most important next step\n - **Files Change",
"cost": {
"context_tokens": 207
}
}
Fetch it by URL: GET /api/v1/registry/abstract-data-texas-turnout-scraper-session-closer-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.