open-responses-agent

open-responses-agent is a cursor rule for Cursor from OthmanAdi/open-responses-agent-skill. It costs 0 tokens per session (1,092 once invoked), scanned A, original, MIT.

A set of development instructions for building autonomous agents with the Open Responses API through Hugging Face's model-routing service. It explains request and response formats, model-provider suffixes, and agent tools.

In plain words
What is it for?
Use it when writing agents that call models, provide instructions and input, invoke tools, select a provider, or process reasoning, tool-call, message, and usage data.
Why use it?
It gives developers a consistent way to send tasks to different model providers and handle agent responses without designing each connection separately.

Cursor rule for Cursor

Written for Cursor: installed under .cursor/.

Install

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.

agentmods
npx agentmods add rules/othmanadi/open-responses-agent-skill/open-responses-agent
Clone the repo
git clone --depth 1 https://github.com/OthmanAdi/open-responses-agent-skill

Made for: Cursor.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for open-responses-agent

README.md
[![agentmods](https://agentmods.dev/badge/rules/othmanadi/open-responses-agent-skill/open-responses-agent.svg)](https://agentmods.dev/rules/othmanadi/open-responses-agent-skill/open-responses-agent)
Your own site
<a href="https://agentmods.dev/rules/othmanadi/open-responses-agent-skill/open-responses-agent"><img src="https://agentmods.dev/badge/rules/othmanadi/open-responses-agent-skill/open-responses-agent.svg" alt="Measured on agentmods" height="20"></a>
Per session 0 Nothing until a file matches its globs; then the whole rule loads.
When invoked 1,092 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00000 $0.01092
Opus 5 $0.00000 $0.00546
Sonnet 5 $0.00000 $0.00218
Haiku 4.5 $0.00000 $0.00109

Measured 5d ago against content hash 5e4842d0a11c, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

open-responses-agent 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 5d ago.

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.

.cursor/rules/open-responses-agent.mdc · 140 lines

How it starts

The opening of the file, as written. The whole thing — 140 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Open Responses Agent Development

You are an expert at building autonomous agents using the Open Responses API via the HuggingFace Inference Providers router.

Core Concepts

1. Single Unified Endpoint

All requests go to: https://router.huggingface.co/v1/responses Provider selection is done via MODEL SUFFIX (e.g., :groq, :together, :nebius, :auto)

2. Request Structure

{
  "model": "moonshotai/Kimi-K2-Instruct-0905:groq",
  "instructions": "You are a helpful assistant.",
  "input": "User's task or request",
  "tools": [...],
  "tool_choice": "auto",
  "reasoning": { "effort": "medium" }
}

3. Response Structure

{
  "id": "resp_abc123",
  "model": "moonshotai/Kimi-K2-Instruct-0905",
  "output": [
    { "type": "reasoning", "content": "Let me think..." },
    { "type": "function_call", "name": "search", "arguments": {...} },
    { "type": "function_call_output", "output": "..." },
    { "type": "message", "content": "Final response" }
  ],
  "output_text": "Final response text (convenience helper)",
  "usage": { "input_tokens": 100, "output_tokens": 200 }
}

Provider Suffixes

Use model suffix to select provider (single endpoint for all):

  • :groq - Groq (fast inference)
  • :together - Together AI (open weight specialist)
  • :nebius - Nebius AI (European infrastructure)
  • :auto - Automatic provider selection

Example: meta-llama/Llama-3.1-70B-Instruct:together

Tool Definition Format

Tools are defined at TOP LEVEL (name, description, parameters) - NOT nested in function:

{
  "type": "function",
  "name": "get_weather",
  "description": "Get weather for a location",
  "parameters": {
    "type": "object",
    "properties": { ... },
    "required": [...]
  }
}

Reasoning Visibility

  • RAW (open weight models): Full content field
  • SUMMARY (some proprietary): summary field with sanitized reasoning
  • ENCRYPTED (most proprietary): encrypted_content field, not accessible

Control reasoning depth with: "reasoning": { "effort": "low" | "medium" | "high" }

Read the full file on GitHub · 140 lines

Changes

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.

  1. 5d ago First seen · 140 lines · 0 tokens per session scan A 5e4842d0a11c

Subscribe to this mod's changes

open-responses-agent is a cursor rule published in the GitHub repository OthmanAdi/open-responses-agent-skill (3 stars, last pushed 7mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,092 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 2026-08-31.