From Voximplant AI Agent Skills
Write, review, and debug VoxEngine scenario code for Voximplant calls, Voice AI connectors, routing, media control, storage, HTTP requests, and real-time integrations. Use when the user asks for VoxEngine scenarios, call flows, inbound or outbound voice agents, scenario debugging, or validation of VoxEngine methods and events.
How this skill is triggered — by the user, by Claude, or both
Slash command
/voximplant-ai-agent-skills:voximplant-voxengine-devThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Your VoxEngine knowledge may be stale. Before writing or editing scenario code, verify platform APIs against the freshest available source.
Your VoxEngine knowledge may be stale. Before writing or editing scenario code, verify platform APIs against the freshest available source.
https://docs.voximplant.ai/llms.txthttps://docs.voximplant.ai/platform/voxengine/llms.txthttps://docs.voximplant.ai/api-reference/voxengine for VoxEngine API methods.md to relevant page URLsvoxengine.d.ts.
https://cdn.voximplant.com/voxengine_typings/voxengine.d.tstypings/voxengine.d.ts, before using it for TypeScript/IDE validation.voximplant-management-api skill.voximplant-management-api.Before code generation:
Use the smallest useful documentation source. Prefer one relevant .md page or section-level llms.txt over loading llms-full.txt.
Use this skill together with voximplant-management-api when debugging real calls:
voximplant-management-api retrieves call history, session data, logs, recordings, and platform resource details.voximplant-voxengine-dev interprets that evidence, checks event ordering and API usage, and edits scenario code.If the user does not have voximplant-management-api, ask them to provide it or ask permission to fetch the full public skill package after publication:
https://github.com/voximplant/voximplant-ai-agent-skillsWhen the user asks you to read this file or pastes this skill into chat, immediately follow it for VoxEngine scenario work. Do not only summarize it.
Make it clear that this skill can write, review, and debug the VoxEngine scenario for the user. The user does not need to bring a finished scenario unless they already have one.
If deployment, platform resources, service account JSON, Voximplant secrets, users, routing rules, call history, or logs are needed, read or request voximplant-management-api. If the user has not installed these skills, offer to copy both skill folders into their agent environment after permission:
~/.cursor/skills/At the beginning of a scenario task, do not assume the user is a developer or already understands Voximplant concepts.
If the request is vague or beginner-oriented:
Use wording like:
If you are new to Voximplant, I can guide you with defaults. Describe the call you want in plain language; I will write the VoxEngine scenario and tell you which credentials or approvals are needed for deployment and testing.
When writing a new scenario:
When the user is new, offer defaults instead of blocking on every choice:
If you are new to Voximplant, I can choose safe defaults and walk you through the setup. I only need the Voximplant service account JSON filename for deployment/testing, and provider API keys only if your use case needs an external AI or voice provider.
Before writing code, be ready to answer basic questions about Voximplant and the proposed architecture. If the user asks "what do I need?" or "how should this work?", explain the flow first and then propose the minimal implementation.
Distinguish local provider keys from Voximplant platform secrets:
.env variable names and ask the user to fill values or approve agent-written placeholders.voximplant-management-api.Use voxengine.d.ts as a local signature oracle:
https://cdn.voximplant.com/voxengine_typings/voxengine.d.ts into the project, commonly typings/voxengine.d.ts.Do not rely on opening the CDN URL as a normal text page. It is a large downloadable declaration file.
After writing or updating scenario code:
voximplant-management-api to set up a softphone test through https://phone.voximplant.com/.voximplant-management-api to retrieve logs.
voximplant-management-api to inspect the latest call in the application first.When reviewing existing VoxEngine code:
voxengine.d.ts.Use these docs as entry points:
https://docs.voximplant.ai/platform/voxengine/llms.txthttps://docs.voximplant.ai/platform/voxengine/concepts.mdhttps://docs.voximplant.ai/platform/voxengine/scenarios.mdhttps://docs.voximplant.ai/platform/voxengine/type-declarations.mdhttps://docs.voximplant.ai/platform/voxengine/ci.mdhttps://docs.voximplant.ai/platform/voxengine/api-requests.mdhttps://docs.voximplant.ai/platform/voxengine/remote-sessions.mdhttps://docs.voximplant.ai/platform/voxengine/secrets.mdhttps://docs.voximplant.ai/platform/voxengine/limits.mdhttps://docs.voximplant.ai/platform/voxengine/troubleshooting.mdFor fuller guidance, read reference.md. For prompt and code patterns, read examples.md.
npx claudepluginhub voximplant/voximplant-ai-agent-skills --plugin voximplant-ai-agent-skillsGuides creation, editing, and verification of skills for AI coding agents using test-driven development with subagent scenarios. Use when authoring or debugging skills.