Skip to content

Permit MCP Gateway documentation#613

Draft
orweis wants to merge 5 commits intomasterfrom
cursor/permit-mcp-gateway-documentation-326e
Draft

Permit MCP Gateway documentation#613
orweis wants to merge 5 commits intomasterfrom
cursor/permit-mcp-gateway-documentation-326e

Conversation

@orweis
Copy link
Contributor

@orweis orweis commented Mar 13, 2026

Rewrites and restructures Permit MCP Gateway documentation to improve clarity, accelerate adoption for developers, and enhance credibility for security teams.

Changes

Landing Page (index.mdx) — Complete Rewrite

  • Outcome-driven hero with dual-audience framing (developers + security/IAM teams)
  • New sections: "Why Teams Adopt", "Use Cases", "When to Use", "Who This Is For"
  • "Built on Permit.io" section explaining the control plane / data plane relationship
  • "Works With Your Existing MCP Stack" compatibility table
  • Deployment options with SaaS (*.agent.security) and customer-controlled

Overview Page (overview.mdx) — Major Rewrite

  • New "How Permit and Permit MCP Gateway Are Connected" section
    • Permit as control plane and default data plane
    • Every Host = Permit Environment (1:1 mapping)
    • Two-dashboard model: app.permit.io + app.agent.security
  • Security and Trust Model section (what the gateway sees/enforces/logs, what it doesn't solve)
  • Comparison table vs API gateways, IAM platforms, manual access

Getting Started (guide.mdx) — Fast Activation Path

  • 5-minute Quick Start at the top (8 steps to first allowed call, denied call, and audit event)
  • Expected results after each setup step
  • Mini troubleshooting block for common quick-start issues

Architecture Page — Executive Summary + Control Plane

  • Executive summary before the deep technical dive
  • New "Permit.io as Control Plane and Data Plane" section with diagram
  • Control plane / data plane framing throughout

Permit.io Integration Page — Connection Explanation

  • New relationship overview table (Permit vs Gateway roles, dashboards, storage)
  • "Every gateway host maps to a Permit environment" explanation
  • Full policy engine capabilities available through the connection

Audit Logs — Investigation Examples

  • 5 practical investigation walkthroughs (denied action, human→agent tracing, server audit, access history, compliance report)

Host Setup — Rollout Guidance

  • 3-phase rollout plan: Pilot → Controlled Expansion → Production
  • Rollout principles (start restrictive, pilot before production, review before broadening)

Advanced Features — Agent Interrogation + Claims Hygiene

  • New: Agent Interrogation — agentic-native identity mechanism using the MCP protocol itself
    • Composite identity: delegating human + workflow context + agent fingerprint
    • Value for policy: hard-gated tool access, per-workflow policy, drift-triggered reactions
    • Value for auditing: persistent identity, declared vs observed intent, full identity chain
    • Value for prompt injection protection: baseline fingerprinting, multi-dimensional drift detection
  • Linked Agent Verification and Intent-Based Access Control to Interrogation as underlying mechanism
  • Feature maturity summary table with honest status indicators
  • Maturity callouts for features under active development
  • Flagged marketing-site claims not yet supported in docs

Naming Consistency — All Files

  • "Agent Security" → "Permit MCP Gateway" across all 12 documentation files
  • Config snippet names updated to permit-mcp-gateway
  • Zero remaining marketing metaphors or inflated claims

cursoragent and others added 3 commits March 13, 2026 15:53
…ateway

- Rewrite index.mdx as outcome-driven landing page with dual-audience framing
- Add 'When to Use', 'Who This Is For', deployment options, and use case sections
- Fix naming: 'Permit MCP Gateway' as primary name, not 'Agent Security'
- Rewrite overview.mdx with Security & Trust Model section
- Add comparison table vs existing tools (API gateways, IAM platforms)
- Add fast 5-minute activation path to Getting Started guide
- Add expected results after each setup step
- Add mini troubleshooting block for quick start

Co-authored-by: Or Weis <[email protected]>
- Add executive summary to architecture page
- Add practical investigation examples to audit logs page
- Add rollout guidance section to host setup page
- Fix 'Agent Security' → 'Permit MCP Gateway' naming across all pages
- Rewrite advanced features with claims hygiene and maturity indicators
- Soften unverified claims (anomaly detection, shadow detection, sub-ms decisions)
- Add feature maturity summary table to advanced features

Co-authored-by: Or Weis <[email protected]>
@cursor
Copy link

cursor bot commented Mar 13, 2026

Cursor Agent can help with this pull request. Just @cursor in comments and I'll start working on changes in this branch.
Learn more about Cursor Agents

@netlify
Copy link

netlify bot commented Mar 13, 2026

Deploy Preview for permitio-docs ready!

Name Link
🔨 Latest commit e73b6f5
🔍 Latest deploy log https://app.netlify.com/projects/permitio-docs/deploys/69b5bb905f0792000797c772
😎 Deploy Preview https://deploy-preview-613--permitio-docs.netlify.app
📱 Preview on mobile
Toggle QR Code...

QR Code

Use your smartphone camera to open QR code link.

To edit notification comments on pull requests, go to your Netlify project configuration.

cursoragent and others added 2 commits March 14, 2026 16:41
- Explain that Permit.io is the control plane and default data plane
- Clarify every Host maps 1:1 to a Permit Environment
- Document the full power of Permit policies available through the connection
- Explain the two-dashboard model: app.permit.io vs app.agent.security
- Add control plane / data plane diagram to architecture page
- Expand permit-integration page with relationship overview table
- Add 'Built on Permit.io' section to landing page
- Update deployment options with SaaS subdomain and PDP details

Co-authored-by: Or Weis <[email protected]>
- Describe agentic-native identity mechanism using the MCP protocol
- Explain composite identity: delegating human + workflow context + fingerprint
- Detail value for policy enforcement: hard-gated tool access, per-workflow
  policy, drift-triggered reactions, step-up consent
- Detail value for auditing: persistent identity, declared vs observed intent,
  full identity chain, drift history
- Detail value for prompt injection protection: baseline fingerprinting,
  multi-dimensional drift detection, session isolation
- Link existing Agent Verification and Intent-Based Access Control sections
  to Agent Interrogation as their underlying mechanism
- Update feature maturity summary table

Co-authored-by: Or Weis <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants