OpenRouter
Official@openrouterteam
The unified interface for LLMs. Find the best models & prices for your prompts.
Agent Skills by OpenRouter
Showing 92 vetted skills indexed across 5 GitHub repositories.
add-benchmark
Implements new benchmarks in the OpenRouter harness with dataset, solver, scorer, and calibration steps.
research-batch-provider
Researches a new batch provider's API and produces a cited capability matrix with live captures and fixtures.
find-negative-balance-users
Query ClickHouse for users whose intraday-corrected balance is currently negative.
e2e-vercel-ai-sdk
Runs Vercel AI SDK end-to-end tests against local or production OpenRouter environments.
db-integration-tests
Write bun:test integration tests for Kysely queries against a real local PostgreSQL database.
stage-endpoint
Stage model endpoints in local Postgres and test them via curl against the local cfw-api.
seed-activity-explorer-data
Seeds deterministic ClickHouse usage data for Activity Explorer verification.
cfw-api-cpu-memory-optimization
Measures cfw-api per-request CPU, memory, and latency using api-perf Cloudflare Worker Previews and Datadog metrics.
dataflow-staging-experiment
Designs and launches A/B experiments on the usage-record Dataflow pipeline against mirrored prod traffic.
cfw-api-startup-optimization
Diagnose and optimize Cloudflare Worker startup CPU time for the cfw-api service.
image-e2e-testing
Runs local end-to-end tests for image-generation providers with response validation and cost reconciliation.
backfill-cf-analytics
Replay Cloudflare GraphQL analytics into Datadog to fill gaps left by sync outages.
video-add-adapter
Implements a video-generation provider adapter with async submit/poll mapping, pricing, and tests.
debug-ci-merge-queue
Diagnose GitHub merge-queue evictions and hanging CI jobs across runner fleets.
frontend-screenshots
Capture screenshots of frontend changes on PRs or branches via Storybook or local dev.
add-router-model
Guides adding a new router model to OpenRouter through enum, plugin, seed, and UI steps.
batch-api-audit
Audits Batch API pull requests against layer boundaries, streaming invariants, and error-handling contracts.
setup-quality-tournament-env
Configures local dev stack, admin access, and evidence capture for quality-tournament verification.
batch-sync-fixtures
Capture real sync-API and provider batch responses into committed JSONL test fixtures.
core-web-vitals-hunt
Diagnose and fix one Core Web Vitals regression using Datadog RUM field data.
detect-proxy-shaped-relay-accounts
Detect accounts reselling frontier models through relays using ClickHouse traffic-shape queries.
tts-adoption-report
Generates daily TTS adoption reports from ClickHouse and Datadog metrics posted to Slack.
audit-slow-db-queries
Audits slow PostgreSQL queries using Datadog DBM metrics, samples, and explain plans.
embeddings-add-adapter
Implements an embeddings provider adapter with multimodal mapping, batch routing, pricing, and tests.
Frequently Asked Questions About OpenRouter
FAQPage SchemaWhat specific tasks can developers perform using these capabilities?▼
Developers can integrate 300+ generative models, execute speech-to-text and text-to-speech conversions, generate and edit images, and manage asynchronous video processing. Additionally, the platform enables querying model benchmarks and retrieving detailed usage analytics for billing and performance monitoring.
Which target personas benefit most from these technical resources?▼
These resources are designed for TypeScript engineers, full-stack developers, and system architects building modular interfaces that require interoperable model access. It is particularly suited for those managing complex model-calling logic, authentication, and data-driven performance tracking within their own software environments.
What are the prerequisites for implementing these integrations?▼
Implementation requires a TypeScript development environment and familiarity with asynchronous request handling. Developers must manage OAuth2 PKCE credentials for secure authentication and utilize the provided type-safe interfaces to interact with model metadata, generation endpoints, and analytics schemas.