How long does the integration take?

Change one base URL in an OpenAI-compatible client, or install the Python or TypeScript SDK; the quickstart routes a first request in a few minutes.

Will this lock us into another vendor?

No. The router is Apache-2.0 open source, your provider keys stay yours, and the API is OpenAI-compatible, so leaving is a change of base URL.

What happens when a provider fails?

The router falls back to the next model your policy allows and records the fallback in the trace, so an outage costs a retry, not an incident.

sylenix-guard - Skill - OpenSmartRoute

Name: sylenix-guard
Author: ANIMANxd

Imported from ANIMANxd/sylenix-guard (AGENTS.md). Install upstream with npx skills add ANIMANxd/sylenix-guard. Copyright stays with the author.

AGENTS.md — Sylenix Guard (POC phase)

Sylenix Guard is a stdio gateway that sits between an AI agent host and one MCP server. This repository is in its 72-hour proof-of-concept phase and contains Sylenix Guard only.

Where things are

POC_SPIKE.md: scope, plan and decision gates. It governs what you may build right now.
docs/SPEC_FULL.md: the long-term product spec. Read its section 2 (axioms) and section 3 (threat model) for intent. Do not implement anything from it that POC_SPIKE.md puts out of scope.
cmd/sylenix-guard/main.go: the whole proxy. One file, standard library only.
demo/: real-agent harness, demo MCP server, protocol tests.
scripts/: helper scripts.

Commands

make check             # gofmt, go vet, protocol tests; must stay green
make build             # builds bin/sylenix-guard
make harness-scripted  # wiring check only; never quote its numbers

Rules

Keep the proxy in cmd/sylenix-guard/main.go. No new packages, no third-party dependencies.
Do not add features beyond POC_SPIKE.md section 2. If something seems needed, say so instead of building it.
Never weaken or delete a test, or loosen a block, to make something pass. Fix the cause.
The guard must never log argument values, secrets or tool output. The harness may record them in per-run transcripts because they are test data.
Unknown means most restrictive: unmapped tools, unparseable frames, anything the guard cannot classify.
Do not claim a behavior the code does not have. Update the limitations in POC_SPIKE.md section 7 instead.
Never reference other Sylenix Labs products in this repository.

Install sylenix-guard 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": "animanxd-sylenix-guard-sylenix-guard", "kind": "skill", "name": "sylenix-guard", "description": "Sylenix Guard is a stdio gateway that sits between an AI agent host and one MCP server. This repository is in its 72-hour proof-of-concept phase and contains Sylenix Guard only.", "publisher": "ANIMANxd", "version": "1.0.0", "capabilities": { "domains": [ "coding" ], "tags": [ "skill-md", "github-agents" ], "languages": [ "en" ] }, "quality_prior": 0.6, "examples": [ "Sylenix Guard is a stdio gateway that sits between an AI agent host and one MCP server. This repository is in its 72-hour proof-of-concept phase and contains Sylenix Guard only." ], "primary": false, "metadata": { "source": { "provider": "github-agents", "repository": "https://github.com/ANIMANxd/sylenix-guard", "path": "AGENTS.md", "ref": "4c2a87e7dbbe2791e2ed9cb8d78e9cc18353ca89", "url": "https://github.com/ANIMANxd/sylenix-guard/blob/4c2a87e7dbbe2791e2ed9cb8d78e9cc18353ca89/AGENTS.md", "key": "ANIMANxd/sylenix-guard/AGENTS.md" } }, "instructions": "# AGENTS.md — Sylenix Guard (POC phase)\n\nSylenix Guard is a stdio gateway that sits between an AI agent host and one MCP server. This repository is in its 72-hour proof-of-concept phase and contains Sylenix Guard only.\n\n## Where things are\n\n- `POC_SPIKE.md`: scope, plan and decision gates. It governs what you may build right now.\n- `docs/SPEC_FULL.md`: the long-term product spec. Read its section 2 (axioms) and section 3 (threat model) for intent. Do not implement anything from it that `POC_SPIKE.md` puts out of scope.\n- `cmd/sylenix-guard/main.go`: the whole proxy. One file, standard library ", "cost": { "context_tokens": 415 } }

sylenix-guard

About

AGENTS.md — Sylenix Guard (POC phase)

Where things are

Commands

Rules

Install sylenix-guard and it becomes one more candidate the router can pick - when it fits.

What it costs to adopt - in engineering time.

One base URL

No vendor lock-in

Failover built in

Use it

Manifest

Reviews