Get listed on OWASP AI Security Solutions Landscape Q2 2026

[pshkv]

Objective

Submit SINT Protocol for inclusion in the OWASP AI Security Solutions Landscape for Agentic AI Q2 2026.

Why

This is where enterprise buyers and researchers discover agentic AI security tools. Microsoft AGT is already listed. SINT's 10/10 OWASP ASI coverage + physical AI focus is a unique differentiator.

Evidence

• Machine-readable conformance: docs/conformance/owasp-asi-mapping.md
• 29 fixture pairs: packages/conformance-tests/fixtures/security/owasp-asi-conformance.v1.json
• Key differentiator: Only framework with physical constraint enforcement in tokens

Priority: P1

[pshkv]

Filed submission:

• Landscape request: Request: Include SINT Protocol in AI Security Solutions Landscape for Agentic AI GenAI-Security-Project/GenAI-Agent-Security-Initiative#11
• ASI03 enforcement comment: Fill ASI03: Identity & Privilege Abuse — credential lifecycle, delegation ceiling, public client pattern GenAI-Security-Project/GenAI-Agent-Security-Initiative#4 (comment)

The submission covers all 10 ASI categories with machine-readable conformance mapping and fixture references. Key differentiator highlighted: only framework with physical constraint enforcement in cryptographic tokens.

Waiting for maintainer response. Will follow up if no response within 7 days.

[pshkv]

Recommended exit criteria for this issue:

• submission is actually sent to the OWASP AI Security Solutions Landscape intake path
• submission cites the repo, docs, and conformance evidence directly
• differentiator is stated in one line: governance for physical and safety-critical agent actions, not just generic agent security
• acceptance or follow-up request is linked back into this thread

This should stay in Now, but it should remain a credibility operation rather than distracting from #128 as the main builder-facing artifact.

[pshkv]

To maximize acceptance odds for OWASP landscape, keep it extremely concrete:

• Product name: SINT MCP (installable MCP proxy)
• Category fit: “Agent Security / Runtime Governance / Tooling Security”
• Proof links (pick 2-3 only):
  • Glama server page
  • interceptor quickstart doc (docs/guides/sint-pdp-interceptor-quickstart.md)
  • evidence receipts + JCS canonicalization in packages/evidence-ledger

If you paste the OWASP landscape submission form fields here (or link), I can draft the exact text per field so it’s copy/paste.

[pshkv]

Here’s copy/paste submission text you can use for the OWASP AI Security Solutions Landscape entry.

Project / Product name
SINT MCP

One-liner
Security-first MCP proxy that enforces capability-token policy on every tool call (T0–T3 tiers), supports approval gates for high-consequence actions, and generates tamper-evident audit receipts.

Category / tags (suggested)
Agent Security, Runtime Governance, Authorization, MCP, Policy Enforcement, Audit

Open source license
Apache-2.0

Repository
https://github.com/sint-ai/sint-protocol

Primary “try it” proof
5-minute interceptor demo: https://github.com/sint-ai/sint-protocol/blob/main/docs/guides/sint-pdp-interceptor-quickstart.md

Install / usage
npx -y sint-mcp

Evidence / credibility links (pick 1–2)

• Glama listing: https://glama.ai/mcp/servers/sint-ai/sint-protocol
• OWASP ASI01–ASI10 conformance mapping + fixtures:
  • https://github.com/sint-ai/sint-protocol/blob/main/docs/conformance/owasp-asi-mapping.md
  • https://github.com/sint-ai/sint-protocol/tree/main/packages/conformance-tests/fixtures/security

Short description (2–3 sentences)
SINT MCP sits between AI agents and external tools/actuators as an enforcement context. Before any tool executes, it validates capability tokens, assigns an approval tier (T0–T3), and can escalate or deny high-risk actions. Every decision can be paired with a verifiable, hash-chained receipt to support audit, compliance, and fail-closed safety behavior.