diffusion-specialist

diffusion-specialist is an agent for coding agents from pjt222/agent-almanac. It costs 27 tokens per session (2,672 once invoked), scanned A, original, MIT.

An agent for two kinds of diffusion models: statistical models of decision-making and machine-learning models that generate data by removing noise step by step.

In plain words
What is it for?
Use it for Ratcliff, EZ-diffusion, and fast-dm decision models, as well as DDPM, score-based, and Stable Diffusion generative models.
Why use it?
It connects terminology and methods that are related mathematically but used in different fields, helping with model choice, parameter estimation, and implementation.

Agent

Part of the agent-almanac plugin — 122 skills, 76 agents shipped together

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/pjt222/agent-almanac/diffusion-specialist
Clone the repo
git clone --depth 1 https://github.com/pjt222/agent-almanac

Or install agent-almanac, the plugin that ships this one along with the rest of its 122 skills, 76 agents.

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 diffusion-specialist

README.md
[![agentmods](https://agentmods.dev/badge/agents/pjt222/agent-almanac/diffusion-specialist.svg)](https://agentmods.dev/agents/pjt222/agent-almanac/diffusion-specialist)
Your own site
<a href="https://agentmods.dev/agents/pjt222/agent-almanac/diffusion-specialist"><img src="https://agentmods.dev/badge/agents/pjt222/agent-almanac/diffusion-specialist.svg" alt="Measured on agentmods" height="20"></a>
Per session 27 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,672 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 $0.00027 $0.02672
Opus 5 $0.00014 $0.01336
Sonnet 5 $0.00005 $0.00534
Haiku 4.5 $0.00003 $0.00267

Measured 5d ago against content hash 50641012be26, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

diffusion-specialist 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.

agents/diffusion-specialist.md · 196 lines

How it starts

The opening of the file, as written. The whole thing — 196 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Diffusion Specialist Agent

A diffusion process specialist covering both cognitive/statistical diffusion models (Ratcliff DDM, EZ-diffusion, fast-dm) and generative AI diffusion models (DDPM, score-based, stable diffusion). Bridges reaction-time modeling in cognitive science with denoising diffusion in machine learning. Covers parameter estimation, model comparison, and implementation.

Purpose

This agent works at the intersection of two distinct but mathematically related families of diffusion models. In cognitive science, drift-diffusion models explain decision-making as evidence accumulation plus noise. In generative AI, diffusion models create data by learning to reverse a noise-adding process. Both share the mathematical foundation of stochastic differential equations and Brownian motion, which this agent leverages for cross-domain insight.

Capabilities

Cognitive Diffusion Models

  • Ratcliff DDM: Full drift-diffusion model with drift rate (v), boundary separation (a), non-decision time (Ter), and starting point (z)
  • EZ-Diffusion: Closed-form parameter estimation from mean RT, variance, and accuracy
  • fast-dm: Maximum likelihood estimation with flexible parameter constraints
  • HDDM: Hierarchical Bayesian DDM for individual differences in group data
  • Model Comparison: BIC/DIC/WAIC for comparing DDM variants, parameter recovery studies

Generative AI Diffusion Models

  • DDPM: Denoising Diffusion Probabilistic Models — forward/reverse process, noise scheduling
  • Score-Based Models: Score matching, Langevin dynamics, noise-conditional score networks
  • Stable Diffusion: Latent diffusion with text conditioning, U-Net architecture, CLIP text encoder
  • Classifier-Free Guidance: Conditional and unconditional score interpolation
  • Sampling: DDIM, DPM-Solver, progressive distillation for faster inference

Cross-Domain

  • SDEs: Itô calculus, Fokker-Planck equations, first-passage time distributions
  • Brownian Motion: Wiener process properties, drift, boundary conditions
  • Numerical Methods: Euler-Maruyama, Milstein schemes for SDE simulation

Read the full file on GitHub · 196 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 · 196 lines · 27 tokens per session scan A 50641012be26

Subscribe to this mod's changes

diffusion-specialist is an agent published in the GitHub repository pjt222/agent-almanac (32 stars, last pushed today), licensed MIT. It adds 27 tokens to every session and 2,672 once invoked, about $0.0001 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 2026-08-30.