ContextPie
ProductHow it worksFeaturesDocsPricing
Sign inGet started
Get started
  • Quickstart
  • Workspaces
  • Authoring a skill
  • Connecting an agent
MCP tools
  • MCP tools
Concepts
  • Progressive disclosure
  • Versioning
  • Analytics
Docs

ContextPie documentation

Author multi-file skills, serve them over one MCP endpoint your agents read at runtime, and measure what got discovered, loaded, and ignored. Start with the quickstart, or jump to any topic below.

Get started

Quickstart
Connect ContextPie to your agent in one step and see your first telemetry.
Workspaces
How ContextPie organizes skills, endpoints, API keys, and people.
Authoring a skill
How to write Agent Skills that agents can actually discover and use.
Connecting an agent
Point any MCP client at the hosted endpoint and the three tools appear.

MCP tools

MCP tools
The three tools agents call — find_skills, get_skill, and get_skill_file.

Concepts

Progressive disclosure
Agents pull detail only as they need it, keeping context lean and cheap.
Versioning
Skills are immutable and versioned — saving creates a new version, agents read current.
Analytics
Turn the MCP traffic you already serve into a map of what agents use, miss, and ignore.
ContextPie

An MCP-native skill registry, AI authoring studio, and analytics layer. A slice of context for every agent.

Product
  • Features
  • Pricing
  • How it works
  • Changelog
Docs
  • Quickstart
  • Authoring a skill
  • Connecting an agent
  • MCP tools
Concepts
  • Progressive disclosure
  • Versioning
  • Analytics
  • All docs
Company
  • About
  • Blog
  • Sign in
© 2026 ContextPie. Working name — building in the open.
PricingDocsContact