Pipeline mutations are fully integrated into the agent workflow.
You can ask the assistant to create agents, connect nodes or restructure flows and every canvas action is streamed live into the UI with audit log and undo support.
One of the biggest changes is the response architecture.
Assistant messages are now streamed as structured envelope parts instead of plain text: markdown blocks, tool execution states, visualization plans, iframe widgets, canvas mutations and error states.
The agent can actively operate inside the workspace.
Read files, inspect audit docs, search vulnerability knowledge bases, call external MCP tools, generate interactive widgets and mutate the pipeline graph directly from chat.
VulnFlow is built on local-first principles.
1. You don't need Claude Opus or ChatGPT-5.4 to write a simple vulnerability report.
2. Sensitive information should stay confidential.
What's the BEST part of VulnFlow - audit tool for smart contracts?
You're not limited to pre-built skills. You can CREATE your own skills & instructions — and compose them however you want.
Here's how to stay ahead of exploit trends 👇
📄 Contract mode: single-file analysis
🗂 Cluster mode: dependency-aware multi-contract grouping
Move from ad-hoc auditing to composable systems where reasoning, rules & context are explicitly defined & controlled.
Workflows are graphs, serializable to JSON, and reusable across projects.
VulnFlow doesn't just run models — it orchestrates a full audit architecture for consistent, scalable security reviews.
With VulnFlow, you define visual workflows where each node = a specific step:
• LLM reasoning
• Rule-based validation
• Data transformation
• External tool interaction
No more manually chaining tools. Build reproducible audit pipelines visually.
The idea sparked while building my own auditing bot.
I kept missing a visual layer — a way to intuitively assemble & manage the entire pipeline from reusable components.










