Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add rules/adaline/gateway/providergit clone --depth 1 https://github.com/adaline/gatewayWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00000 | $0.01277 |
| Opus 5 | $0.00000 | $0.00639 |
| Sonnet 5 | $0.00000 | $0.00255 |
| Haiku 4.5 | $0.00000 | $0.00128 |
Grade A, and why
provider scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 161 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Provider Package Rules
Package Overview
The provider package defines the core interfaces and base classes that all AI providers must implement. It provides a unified abstraction layer for different AI services.
Provider Interface Rules
1. ProviderV1 Interface
- ALWAYS implement the
ProviderV1interface for new providers - ALWAYS use generic constraints for configuration types
- ALWAYS provide both chat and embedding model support
- ALWAYS implement version tracking with
readonly version = "v1" - ALWAYS use descriptive provider names
2. Model Factory Methods
- ALWAYS implement
chatModelLiterals()returning available model names - ALWAYS implement
chatModelSchemas()returning Zod schemas - ALWAYS implement
chatModel(options)returning configured instances - ALWAYS implement
embeddingModelLiterals()returning available model names - ALWAYS implement
embeddingModelSchemas()returning Zod schemas - ALWAYS implement
embeddingModel(options)returning configured instances
Model Implementation Rules
1. Chat Model Implementation
- ALWAYS implement the
ChatModelV1interface - ALWAYS provide model schema with name, provider, and capabilities
- ALWAYS implement
completeChat()method for synchronous responses - ALWAYS implement
streamChat()method for streaming responses - ALWAYS implement
getModelPricing()for cost calculation - ALWAYS implement
getCompleteChatUrl(),getCompleteChatHeaders(),getCompleteChatData() - ALWAYS implement
transformCompleteChatResponse()for response normalization
2. Embedding Model Implementation
- ALWAYS implement the
EmbeddingModelV1interface - ALWAYS provide model schema with name, provider, and capabilities
- ALWAYS implement
getEmbeddings()method - ALWAYS implement
getEmbeddingsUrl(),getEmbeddingsHeaders(),getEmbeddingsData() - ALWAYS implement
transformEmbeddingsResponse()for response normalization
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 161 lines · 0 tokens per session scan A 3a9ec3f74d9e
provider is a cursor rule published in the GitHub repository adaline/gateway (605 stars, last pushed 1mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,277 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other cursor rules, from other repositories
js-sdk-genai-framework-instrumentation
Guide for implementing JS SDK agent and framework instrumentations, such as LangChain, LangGraph, OpenAI Agents, Claude Agent SDK, Google ADK, and Vercel AI.
coding-agents-hook
Authoring and testing coding-agent hook adapters, including payload probing, session state, capture modes, and plugin manifests. Consult the telemetry convention when changing canonical attributes.
coding-agents-convention
Canonical coding-agent telemetry schema, inbound mappings, and data contracts. Applies to semantic conventions, emission, processing, and readers—not hook implementation mechanics.
security-testing
Authorization, input validation, CE boundary, secrets, and test guidance for OpenLIT server code.
documentation
Standards for OpenLIT documentation, MDX structure, navigation, examples, and terminology.
js-sdk-genai-instrumentation
Guide for aligning JS SDK GenAI provider instrumentations with OTel semantic conventions and the Python SDK reference implementation.