github.com/cheaper-inference/OmniRoute83.6k
Never stop coding.
npx skills add cheaper-inference/OmniRoute -g -y
Sends tasks to the OmniRoute agent-to-agent server and tests its JSON-RPC protocol.
Creates, schedules, syncs and restores OmniRoute backups from the command line.
Submits and monitors batch inference jobs and their files from the command line.
Sends chat completions, streams replies and runs an interactive chat session from the terminal.
Configures and tests prompt compression rules and previews compressed output.
Manages context configurations, filter sets and conversation sessions from the terminal.
Shows token usage and cost breakdowns by provider, model or date and exports reports.
Creates and runs eval suites, compares model scores and adds eval runs to CI.
Checks server health, component status and circuit breakers with a live watch view.
Creates, lists, rotates and revokes API keys and manages provider OAuth flows.
Inspects MCP server status, tools and scopes, runs tool calls and reads audit logs.
Lists models and aliases across providers, filtering by provider or capability.
Installs, tests and removes skills and plugins, and manages persistent agent memory.
Reads audit logs, access policies, telemetry and request history for compliance reviews.
Adds, tests and rotates keys for AI provider connections and shows their metrics.
Inspects and resets circuit breakers, cooldowns and quota limits for stuck providers.
Creates, tests and updates routing combos and fallback chains from the command line.
Starts, stops and restarts the OmniRoute server, including daemon mode and port settings.
Runs first-time setup, global CLI settings, environment variables and autostart.
Finds agent skills on GitHub that match installed coding tools and installs them there.
Starts and stops ngrok or Cloudflare tunnels and checks that external URLs are reachable.
Configures the Codex CLI to use OmniRoute as its backend on any major OS.
Calls OmniRoute's six built-in agent skills over the JSON-RPC agent-to-agent protocol.
Creates and rotates API keys and sets scopes, spending limits and expiry dates.
Authenticates API requests with bearer tokens and session cookies, and sets login rules.
Sets spending limits, token quotas and rate limits per key or globally, and tracks use.
Views cache statistics, clears entries and sets TTL and semantic similarity thresholds.
Lists, configures and invokes CLI tool plugins that extend OmniRoute automation.
Creates routing combos with 19 strategies and fallback chains, then tests outcomes.
Configures command, prose and stacked compression, reducing token use by 60 to 90 percent.
Configures RTK filters and context rules, tests compression on samples and manages pipelines.
Triggers backups, restores from archives and manages SQLite export, import and snapshots.
Finds, scores and scans agent skills on GitHub for malware and secrets before installing.
Calls chat, embedding, image, speech, moderation and rerank endpoints through one API.
Connects to the OmniRoute MCP server and its 110 tools across routing, cache and audit.
Lists models across providers, resolves aliases and browses provider-specific variants.
Adds, updates, removes and tests AI provider integrations through the REST API.
Sets HTTP, HTTPS and SOCKS proxies for provider requests, with rotation and tests.
Monitors provider health, circuit breakers, latency percentiles and budget alerts.
Reads and updates system prompts, thinking budgets, IP filters and login requirements.
Syncs OmniRoute configuration and provider connections to and from cloud storage.
Creates ngrok, Cloudflare or custom tunnels to share OmniRoute with remote agents.
Filters call logs and token usage by provider, model, time and cost, and exports them.
Installs, starts, stops and updates embedded services and reads their logs.
Registers and tests webhook endpoints for request and quota events, with delivery retries.
Favors the simplest working solution: reuse existing code and the standard library first.
