Server facts
portkey-admin-mcp
Project
- Author
- s-b-e-n-s-o-n
- License
- Not specified
- Category
- MCP Gateways & Proxies
Tech
- Runtime / language
- Not specified
- Transports
- stdio
- Remote available
- No
- Authentication
- Not specified
Activity
- GitHub stars
- 6
- Last commit
- Data as of
Tags
- Observability
- admin api
- ai gateway
- analytics
- api keys
- configs
- integrations
- llm
- portkey
- prompts
Overview
Prompt lifecycle management anchors portkey-admin-mcp’s access to the Portkey AI Gateway Admin API. Teams can organize templates into collections, create revisions, inspect rendered messages without invoking a model, and publish selected versions. Separate tools execute billable completions, validate billing metadata, migrate prompts through idempotent CI/CD workflows, and copy them between environments. Gateway configuration tools manage routing, caching, retries, and targets, while credential tools distinguish Portkey authentication keys from virtual keys holding provider credentials. Workspace membership, guardrails, throttling policies, and cumulative usage budgets are also manageable.
For MCP administration, the project registers external integrations and server instances, checks connectivity, and controls exposed capabilities. Workspace permissions apply at the integration level, while server tools provide individual user access controls. Analytics cover spending, tokens, latency, errors, caching, and feedback, with breakdowns by user, model, or metadata. Export workflows separate defining a log export from starting processing and retrieving its completed download. Analytics and certain administrative operations are Enterprise-gated. API key creation exposes the secret in the tool response, making it visible in MCP transcripts and model context.
portkey-admin-mcp setup notes
- portkey-admin-mcp records stdio only; confirm a network transport upstream before planning remote access.
- portkey-admin-mcp has no recorded license; confirm reuse terms in its repository.