Back to directory
AI & Automationhawkeyexl/inference
@hawkeyexl/inference
Shared TypeScript LLM inference layer: schema-constrained completion across Anthropic, OpenAI-compatible, Claude CLI, and in-process local llama.cpp providers, with caching, cost accounting, and an LLM-as-judge ensemble.
Suggested install command
npx skills add hawkeyexl/inference/@hawkeyexl/inferenceAlways inspect the linked repository and skill instructions before running commands. Skills are instructions; permissions and execution still matter.
Compatibility
Agent support matrix
3 supported
| Agent | Status |
|---|---|
| Claude Code | Supported |
| OpenCode | Not listed |
| Cursor | Supported |
| MCP | Not listed |
| GitHub Copilot | Not listed |
| Windsurf |