Route private inference requests by task type via Speakeasy endpoints.
npx skills add https://github.com/wrensignal/wrenOS --skill speakeasy
Please help me install this Agent Skill. Skill: speakeasy Source: https://github.com/wrensignal/wrenOS/tree/main/vendor/skills/speakeasy Command: npx skills add https://github.com/wrensignal/wrenOS --skill speakeasy
Enables operators to securely route private AI inference requests through Speakeasy endpoints, enabling auditable, cost-aware workflows without exposing sensitive models or data.
Install and initialize the Speakeasy tool and start routing private inferences with automated wallet handling.
High-intent search queries and answers about installing and using this skill.
Private inference routing directs AI requests through secure endpoints based on task categories like research, deep_think, or codegen. Speakeasy enables this by applying configurable routing rules to operator workflows, ensuring cost-aware and auditable request handling.
Wallet handling for paid inference calls requires generating a hot wallet on first use. Speakeasy automates this process by creating a wallet during the initial request and reusing it for all subsequent paid-path operations.
Private inference routing requires the environment variables SPEAKEASY_BASE_URL and SPEAKEASY_API_KEY to be set. These configure the secure endpoints needed for routing AI requests through the operator workflow.
When private inference endpoints are unavailable, degraded-mode safeguards activate to maintain operation continuity. Speakeasy implements explicit fallback behavior with recoverable paper-mode operation to prevent complete workflow failure.
You can check inference costs and available models using the operator-facing tools speakeasy_balance and speakeasy_models. These tool contracts expose balance checking and model listing capabilities directly within the workflow.
lyticsConfigure external connections, auth providers, and credentials for secure data exchange.
grahama1970Provision, manage, and terminate RunPod GPU instances for LLM workflows.
etylsarinStandardize NX monorepo commands, targets, and generators across projects.
ax128Deploy AegisGate via Docker or local bootstrap and configure token routing.
