Feature ideas

Integrating an Audit Log API into a Model Context Protocol (MCP) server so that Audit logs can be reviewed with MCP and Agents

We would like to use Port as an mcp gateway specifically for JIRA. But to do so we need to dynamically set the header in each request with the currently logged in user. (It doesn't play nice with per user oauth but can auth as a single app and do impersonation) so ideally we want to be able to set xoauth_requestor_id with “ michael.example@domain.com ” as an example.

Bring your own MCP. Port’s AI today only sees what’s inside Port. By bringing in a customer's MCP Server and their AI Agents, they are already pulling in external context from Confluence, Jira, and other systems. Their AI Agents may have deep context surrounding their business. Customer wants Port to be able to tap into that external intelligence so the AI inside Port can act with full awareness of both Port and non-Port data.

While this is motivated primarily by my company being a google cloud shop internally, I believe this is also important for simply having the best range of options as models continue to expand in capabilities.

Goal: Discover, view, and govern all agents, skills, prompts, commands across the organization with hierarchical scopes and approval workflows so that you can standardize AI tooling and prevent agentic sprawl. Business Outcome: Single source of truth for AI entities; eliminate duplicates and shadow; enforce quality standards before distribution.

Goal: Receive AI related entities (prompts, skills, commands, rules, hooks) automatically to my workstation (Cursor, VSCode, Claude Code) so that I always have the latest company and team standards without manual action. Business Outcome: 100% adoption of critical skills and commands (security, compliance); reduced friction for developers.

I just need a way to set a custom URL, and set what model I want to use. We're effectively using bedrock, however we want to be able to control credit usage on a per user basis and potentially charge those to their relevant AWS accounts, hence wanting to use a proxy.

Enhance Port’s GitHub integration to ingest and model GitHub Copilot seat assignments and Copilot usage/engagement metrics (license usage), enabling customers to build accurate adoption + cost efficiency dashboards in Port and reducing reliance on Copilot API endpoints that are being deprecated/sunset. Seat visibility: who has Copilot, when they were last active, seat state, and assignment source. Usage visibility: adoption and engagement metrics over time (org/enterprise), to justify spend and guide enablement. Future-proof integrations: GitHub has deprecated older/beta Copilot usage endpoints and is steering users to newer metrics APIs.