Skip to content

CODESPECT-security/canton-ai-review

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 
 
 
 
 
 
 

Repository files navigation

daml-audit skill

A multi-expert security audit skill for Daml smart contracts on the Canton Network.

What it does

Triggers three independent analysis passes over Daml code — an Authorization & State auditor, a Privacy & Canton auditor, and a skeptical Triager — then merges validated findings into a structured report. The triager round actively challenges each finding to eliminate false positives before anything reaches the final report.

Reference material

File Purpose
reference/daml-checks.md Daml-specific vulnerability checklist (signatories, choices, proposals)
reference/canton-checks.md Canton infrastructure, privacy, and operational concerns
reference/finding-format.md Standardized finding template and severity definitions
reference/multi-expert.md Full three-round verification protocol

Installation

Copy this directory into your Claude Code skills folder:

cp -r daml_review ~/.claude/skills/daml-audit

Disclaimer

This tool uses AI to assist with security analysis. AI is non-deterministic and will not catch every vulnerability — findings may be incomplete, incorrect, or miss context that a human auditor would recognize. Do not rely on this tool as your sole security review before deploying to production.

We strongly recommend an independent professional security review of any Daml contracts handling real assets or sensitive data. CODESPECT specializes in smart contract security audits and can provide that assurance.

Acknowledgments

The skill architecture — multi-expert agents, independent analysis passes, and a triager validation round — was inspired by the agent patterns at forefy.

About

AI-powered security audit skill for Daml smart contracts on Canton

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors