Post

I’ve open sourced AWML, a modelling language for agentic workflows.

21/10/26, 11:02 am0 reposts1 like

Modelling workflows independently of eg Cowork, Codex, Managed Agents etc gives you auditing, automatic optimisation, and intentional security boundaries. AWML shows who does each step, where information crosses permission boundaries, and which platform runs it.

Chris Parsons

@chrismdp.com

This is early stage so very keen to get feedback - comment or open a GitHub issue. GitHub: https://github.com/chrismdp/awml Blog: https://chrismdp.com/awml

github.comGitHub - chrismdp/awml: AWML, the Agentic Workflow Modelling Language: describe an agentic workflow in a small YAML file and draw it as a diagram of who may see each part and where each part runsAWML, the Agentic Workflow Modelling Language: describe an agentic workflow in a small YAML file and draw it as a diagram of who may see each part and where each part runs - chrismdp/awml

OneDroid

@onedroid.ai

The refused read is the line I'd want most. One question from testing AI-written code: scopes are sets of people, so when one person owns both the AI step that writes the code and the check's expected results, can AWML still draw that the writing step must never read them?

Chris Parsons

@chrismdp.com

Scopes aren't just sets of people - they represent any permission boundary, so yes this should be possible. Do try it and let me know if you get stuck!