Key facts
| API compatibility | OpenAI-compatible /v1/chat/completions; drop-in base URL change |
| Models | 30+ models behind one API; open-weight options for private deployment |
| Typical patterns | Ticket triage, RCA drafts, runbooks, code review, SOW drafts |
| Integration path | Connects to ITSM, RMM, code repositories and CRM through middleware |
| Pricing model | Flat monthly self-serve plans with unlimited fair-use usage; no per-token billing — see the live pricing page |
| Free tier | plugsky-micro and plugsky-lite on the free plan, no card required |
| Deployment | Plugsky cloud, your VPC, on-prem and air-gapped options |
| Identity and audit | SSO/SCIM, role-based access, SIEM audit log export and BYOK via KMS or HSM |
TL;DR
- Draft triage, RCA and runbooks for engineers to approve.
- Isolate keys, indexes and audit trails per client tenant.
- Keep credentials and secrets out of prompts entirely.
- Match residency and deployment to each client contract.
- Start free with plugsky-micro and plugsky-lite, no card required.
How it works, step by step
- Pick an internal workflow, such as ticket summarisation or RCA drafts.
- Design tenant isolation for keys, projects and retrieval indexes.
- Keep credentials, secrets and unnecessary client data out of prompts.
- Build against the OpenAI-compatible endpoint with per-tenant keys.
- Configure residency and audit export per client contract.
- Pilot internally, measure handling time, then offer it as a client service.
Try it yourself
Where an AI API fits in IT services
Service providers bill for judgement and speed. The API helps most where output is a draft a senior engineer reviews:
- Ticket triage and summaries: classify and condense incidents so engineers start with context, not a queue.
- RCA drafts: turn timeline notes and logs into a structured root-cause draft the engineer corrects.
- Runbook generation: draft procedures from resolved tickets for review and standardisation.
- Code review assistance: flag likely defects and missing tests for human reviewers to confirm.
- Proposal and SOW drafts: assemble scopes from service catalogues and prior engagements.
Security, privacy and data handling
One provider plus many clients equals a multi-tenant AI architecture. Isolation is the product requirement:
- Separate keys, projects and retrieval indexes per client tenant.
- Keep credentials and secrets out of prompts; the model proposes actions, engineers execute.
- Give each tenant its own audit view so access questions are answerable.
- Match deployment and residency to each client contract, not just your own preference.
Deployment options and model choice
Serve clients from one platform with per-tenant keys and indexes, and move to private deployment for contracts that require it. The same OpenAI-compatible API runs across Plugsky cloud, a private endpoint in your VPC, on-prem and air-gapped, with region-locked planes for residency. One key reaches 30+ models, including open-weight options for offline deployment, and migration is a base URL change. Chat, streaming, JSON mode, function calling, embeddings, RAG and agents are live; audio, images, moderation, files, batch, fine-tuning, assistants and the responses API remain coming soon. The free plan includes plugsky-micro and plugsky-lite with no card, and a 14-day full-access trial covers paid tiers — see the live pricing page for current plans.
From pilot to production
Cross-tenant leakage is the existential risk. Most incidents come from shared infrastructure:
- One shared key or index for every client.
- Sending production credentials or config into prompts.
- No per-tenant audit trail when a client asks who accessed what.
- Assuming one client's residency requirements cover all clients.
- Automating client-facing updates without an engineer's approval.
Start with internal engineering workflows, build tenant isolation into keys and retrieval from day one, and give each client an audit export. Then productise the same pattern across your client base.
Honest comparison
| Capability | Plugsky | Typical per-token API | Building in-house |
|---|---|---|---|
| API compatibility | OpenAI-compatible chat, embeddings and tools | Usually compatible | Full rewrite |
| Deployment | Cloud, VPC, on-prem and air-gapped | Mostly cloud-only | You operate GPUs and serving |
| Data residency | Region selection and sovereign options | Limited regions | You control fully |
| Pricing | Flat monthly self-serve, fair-use usage | Per-token, harder to forecast | GPU plus operations cost |
| Model choice | 30+ models behind one API | Varies by provider | You host every model |
| Industry fit | Ticket triage, RCA drafts, runbooks, code review, SOW drafts | Generic API, you adapt it | You build every workflow |
Frequently asked questions
Can we keep our existing OpenAI SDK code?
Yes. Plugsky exposes an OpenAI-compatible API, so you change the base URL and model name and keep your integration.
Is there a free plan?
Yes. The free plan includes two free models, plugsky-micro and plugsky-lite, and does not require a credit card.
How do we keep client data isolated?
Use separate projects and keys per client, separate retrieval indexes, and per-tenant audit logs. Isolation belongs in the architecture, not just the policy.
Can we resell this to clients?
Plugsky offers a white-label AI API option. Check the terms and packaging with the enterprise team before reselling.
How does pricing work?
Self-serve plans are flat monthly with unlimited fair-use usage; enterprise and white-label agreements cover capacity and SLA terms. See the live pricing page for current plans.
Which endpoints are live today?
Chat completions, streaming, JSON mode, function calling, embeddings, RAG and agents are live. Audio, images, moderation, files, batch, fine-tuning, assistants and the responses API are coming soon.
Can it propose remediation actions?
It can draft proposals, but keep execution behind human approval and your existing change-control process.