agents-note

A Python interface for creating and coordinating AI agents while choosing among several agent frameworks, such as LangChain or AutoGen. It provides the same basic controls even when you change frameworks.

In plain words
What is it for?
Use it to create agents, build workflows, and run or stream their results from Python or a Claude Code conversation.
Why use it?
It removes the need to rewrite your agent code when changing frameworks. It also gives agents and multi-step workflows a consistent way to run asynchronously.

Agent

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 agents/smart-ai-memory/attune-ai/note
Clone the repo
git clone --depth 1 https://github.com/Smart-AI-Memory/attune-ai
Per session 2 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,262 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 $0.00002 $0.01262
Opus 5 $0.00001 $0.00631
Sonnet 5 $0.00000 $0.00252
Haiku 4.5 $0.00000 $0.00126

Measured 2d ago against content hash 9bc135a2171f, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agents-note 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 2d 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.

.help/templates/agents/note.md · 108 lines

How it starts

The opening of the file, as written. The whole thing — 108 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Universal Agent Factory — create, run, and orchestrate AI agents across frameworks

Overview

The agents feature is Attune's Universal Agent Factory — one interface for creating, running, and orchestrating AI agents, backed by your choice of framework (native, LangChain, LangGraph, AutoGen, or Haystack) without rewriting code when you switch frameworks. The entry point is AgentFactory: it picks a framework adapter, and its create_agent / create_workflow methods return BaseAgent / BaseWorkflow objects with a uniform interface.

The agent and workflow run methods (invoke, run, stream) are asyncawait them.

You reach it two ways:

  • the Python API — from attune.agent_factory import AgentFactory, Framework (the primary surface, documented throughout);
  • the /agent skill, inside a Claude Code conversation — create and manage custom agents and teams.

There is no attune agent CLI command and no MCP tool.

Scope. This feature is the framework-agnostic Agent Factory (src/attune/agent_factory/). The release-readiness agent team (src/attune/agents/release/) is documented under release-prep, and the agent state/recovery store (src/attune/agents/state/) is that team's persistence layer — not part of the Factory's public surface.

Concepts

One factory, many frameworks

AgentFactory(framework=None, provider="anthropic", api_key=None, use_case="general") is the entry point. framework is a Framework enum (or its string) — native (the default when unset), langchain, langgraph, autogen, or haystack. Each non-native framework is an optional dependency loaded lazily; AgentFactory.list_frameworks( installed_only=True) reports what's available and AgentFactory.recommend_framework(use_case) suggests one. Call switch_framework(framework) to move an existing factory to another backend.

Create agents and workflows

Method Returns What it does
create_agent(name, role=AgentRole.CUSTOM, model_tier="capable", ...) BaseAgent Build one agent. Many options — capabilities, tools, system_prompt, temperature, memory_enabled, resilience_enabled, …
create_workflow(name, agents, mode="sequential", ...) BaseWorkflow Coordinate several agents (sequential or other modes).
create_tool(name, description, func, args_schema=None) tool Wrap a Python callable as an agent tool.
create_coordinator / create_researcher / create_writer / create_reviewer / create_debugger BaseAgent Role-preset agent shortcuts.
create_code_review_pipeline() / create_research_pipeline(topic, include_reviewer=True) BaseWorkflow Ready-made multi-agent pipelines.
get_agent(name) / list_agents() BaseAgent | None / list[str] Look up agents the factory has created.

Read the full file on GitHub · 108 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. 2d ago First seen · 108 lines · 2 tokens per session scan A 9bc135a2171f

Subscribe to this mod's changes

agents-note is an agent published in the GitHub repository Smart-AI-Memory/attune-ai (10 stars, last pushed 2d ago), licensed Apache-2.0. It adds 2 tokens to every session and 1,262 once invoked, about $0.0000 per session on Opus 5. 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.