Skip to content
This repository was archived by the owner on May 1, 2026. It is now read-only.

Add MOSS to Tools section (AI Agent Security)#43

Closed
Lechatdiabolique wants to merge 2 commits into
arainho:masterfrom
Lechatdiabolique:add-moss
Closed

Add MOSS to Tools section (AI Agent Security)#43
Lechatdiabolique wants to merge 2 commits into
arainho:masterfrom
Lechatdiabolique:add-moss

Conversation

@Lechatdiabolique
Copy link
Copy Markdown

@Lechatdiabolique Lechatdiabolique commented Feb 19, 2026

What is MOSS?

MOSS provides cryptographic signing for AI agent API calls using ML-DSA-44 post-quantum signatures.

Section

Tools > AI Agent Security (new subsection)

Why it belongs here

As AI agents increasingly make API calls autonomously, security teams need:

  • Attribution: Cryptographic proof of which AI agent made which API call
  • Tamper-proof audit trails: Signed records of all agent API interactions
  • Policy enforcement: Block/hold risky API calls before they execute
  • Compliance: Evidence for regulatory requirements

MOSS signs every API call made by AI agents, creating a verifiable chain of custody.

Features

  • ML-DSA-44 post-quantum signatures (NIST FIPS 204)
  • Offline verification without network calls
  • Framework integrations: LangChain, LangGraph, CrewAI, AutoGen
  • Multi-language SDKs: Python, TypeScript, Rust, Java, C#/.NET

Links

@arainho
Copy link
Copy Markdown
Owner

arainho commented May 1, 2026

I appreciate the contribution; however, this resource is out of scope and the MR will be closed.

This repository focuses on open-source tools and community resources directly related to API security. Commercial or freemium services are out of scope for this list, as stated in the contribution guidelines: awesome-api-security contributions.

@arainho arainho closed this May 1, 2026
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants