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.
context-loop - Skill - OpenSmartRoute
Skillv1.0.0
context-loop
Run a bounded maker-checker loop for catalog changes with explicit budgets, gates, and human approval points.
Imported from RobGonWin/deving-research-ecosystem (.agents/skills/context-loop/SKILL.md). Install upstream with npx skills add RobGonWin/deving-research-ecosystem --skill context-loop. Copyright stays with the author.
Context loop
Use LOOP.md, STATE.md, and gate.yaml as the controlling files.
State one narrow objective and list affected paths.
Set iteration, elapsed-time, changed-file, and external-call budgets.
Maker applies the smallest reversible change.
Checker inspects the exact proposed head and runs every command in gate.yaml.
Stop on a failed boundary, unresolved license, budget exhaustion, or human gate.
Record status, public file paths, commands, and pass/fail outcomes in loop/runs.jsonl.
Keep automation at L1 report-only. Never record hidden reasoning, protected inputs, credentials, or copied upstream instructions in the run log.
Use it
Copy one of these into your project. Installing also returns the manifest and these snippets.
yaml
targets:
- https://api.opensmartroute.ai/api/v1/registry/robgonwin-deving-research-ecosystem-context-loop/manifest # or paste the manifest below
Manifest
An Open Capability Manifest: the router reads it to know what this does, what it costs and when to pick it.
{
"ocm": "1",
"id": "robgonwin-deving-research-ecosystem-context-loop",
"kind": "skill",
"name": "context-loop",
"description": "Run a bounded maker-checker loop for catalog changes with explicit budgets, gates, and human approval points.",
"publisher": "RobGonWin",
"version": "1.0.0",
"capabilities": {
"domains": [
"general"
],
"tags": [
"skill-md",
"github"
],
"languages": [
"en"
]
},
"quality_prior": 0.6,
"examples": [
"Run a bounded maker-checker loop for catalog changes with explicit budgets, gates, and human approval points."
],
"primary": false,
"metadata": {
"source": {
"provider": "github",
"repository": "https://github.com/RobGonWin/deving-research-ecosystem",
"path": ".agents/skills/context-loop/SKILL.md",
"ref": "eefd00ddc099f9ad9f1a2ce8e4bdae0fcad58ece",
"url": "https://github.com/RobGonWin/deving-research-ecosystem/blob/eefd00ddc099f9ad9f1a2ce8e4bdae0fcad58ece/.agents/skills/context-loop/SKILL.md",
"key": "RobGonWin/deving-research-ecosystem/.agents/skills/context-loop/SKILL.md"
}
},
"instructions": "# Context loop\n\nUse `LOOP.md`, `STATE.md`, and `gate.yaml` as the controlling files.\n\n1. State one narrow objective and list affected paths.\n2. Set iteration, elapsed-time, changed-file, and external-call budgets.\n3. Maker applies the smallest reversible change.\n4. Checker inspects the exact proposed head and runs every command in `gate.yaml`.\n5. Stop on a failed boundary, unresolved license, budget exhaustion, or human gate.\n6. Record status, public file paths, commands, and pass/fail outcomes in `loop/runs.jsonl`.\n\nKeep automation at L1 report-only. Never record hidden reasoning, protected i",
"cost": {
"context_tokens": 166
}
}
Fetch it by URL: GET /api/v1/registry/robgonwin-deving-research-ecosystem-context-loop/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.
