tech.xns/relayer
5 changes recorded. MCP Harbor re-reads this listing from the official registry every six hours and asks the live server what it exposes, and keeps what changed between one look and the next.
− - Your endpoint is your own Relayer — there is no shared XNS hostname to point at. You self-host the Relayer (see the MCP setup above) and reach it at a domain you control, with your own TLS cert. Use that URL as your S3 endpoint; the examples below use https://relayer.example.com — substitute your own.+ - Your endpoint is your own Relayer — there is no shared production XNS hostname to point at. (The sandbox at https://xns.tech/demo gives each visitor a temporary Relayer with its own keys; it closes after at most 120 minutes and stores nothing, so do not use its endpoint for real data.) You self-host the Relayer (see the MCP setup above) and reach it at a domain you control, with your own TLS cert. Use that URL as your S3 endpoint; the examples below use https://relayer.example.com — substitute+ - Try a live sandbox Relayer (no install, per-visitor, nothing stored): https://xns.tech/demo
− - Independently rated: Glama scores the server tier A — 4.4/5 on tool-definition quality and 5/5 on server coherence across disambiguation, naming consistency, tool count, and lifecycle completeness. https://glama.ai/mcp/servers/xns-cloud/relayer-mcp− - First-party CLI: the xns CLI (built on s5cmd) — `xns auth login --endpoint https://relayer.example.com`, then `xns ls/cp/sync/rm/mv/presign`. Scriptable, `--json` output, light on agent context.− - Reference workflows you can run against XNS storage are being prepared. This section will list each one here, by name and link, once it publishes.− - Video ingestion pipeline (upload, transcode, transcribe): coming soon.− - Document RAG parser (ingest, chunk, embed, store): coming soon.− - No template repositories are published yet. Do not infer or construct a URL for any of the above — treat a template as available only once this section carries its link.+ - Reproducible install: replace @latest with an exact version — npx @xns-cloud/relayer-mcp@<version>. List published versions with: npm view @xns-cloud/relayer-mcp versions+ - Independently rated: Glama scored the server tier A on 2026-08-09 — 4.4/5 on tool-definition quality and 5/5 on server coherence across disambiguation, naming consistency, tool count, and lifecycle completeness. https://glama.ai/mcp/servers/xns-cloud/relayer-mcp+ - First-party CLI: the xns CLI (built on s5cmd) — `xns auth login --endpoint <your-endpoint>` (for example https://relayer.example.com), then `xns ls/cp/sync/rm/mv/presign`. Scriptable, `--json` output, light on agent context.+ - Working recipes for AI pipelines on XNS are published at https://github.com/xns-cloud/xns-ai-cookbooks — single-process starter recipes, each stating where it stops being enough.+ - Multimodal RAG (video speech and frames → transcripts and captions cached in XNS → query): https://github.com/xns-cloud/xns-ai-cookbooks/tree/main/multimodal-rag+ - Agentic document parsing (PDFs and spreadsheets → structured JSON, cached per document): https://github.com/xns-cloud/xns-ai-cookbooks/tree/main/agentic-doc-parsing+ - Fine-tune checkpointing (model weights to and from GPU hosts via S3 multipart): https://github.com/xns-cloud/xns-ai-cookbooks/tree/main/finetune-checkpointing+ - Agent workspace (CrewAI agents exchanging artifacts through a shared bucket): https://github.com/xns-cloud/xns-ai-cookbooks/tree/main/agent-workspace+ - Persistent index (a Qdrant collection snapshotted to XNS and restored without re-embedding): https://github.com/xns-cloud/xns-ai-cookbooks/tree/main/persistent-index+ - This section lists every published template. Do not infer or construct a URL for a template that is not linked here.
We can't find the internet
Attempting to reconnect
Something went wrong!
Attempting to reconnect