KyaniteLabs/Epoch

Overview

KyaniteLabs/Epoch is an MCP server that provides time estimation capabilities for AI agents. It implements multiple estimation models including PERT, COCOMO II, Monte Carlo simulation, sprint forecasting, and token-to-time mapping for cost and schedule risk analysis.

Best for

Best for
Developers building AI agents that need automated time and cost estimation for tasks or sprints

Use cases

• Estimating task durations for AI-driven project planning
• Forecasting sprint timelines and resource allocation
• Mapping token usage to time and cost estimates

How to use

Install

npx tsx scripts/backtest-pert-correction.mjs

Tools exposed

• EPOCH_TRANSPORT
• EPOCH_PORT
• EPOCH_HOST
• EPOCH_DATA_DIR
• EPOCH_COMMUNITY_DIR
• EPOCH_RATE_LIMIT
• EPOCH_SOURCE
• EPOCH_TELEMETRY
• EPOCH_TELEMETRY_ENDPOINT

Tested with

Claude Code, Cursor, Windsurf, VS Code, ChatGPT

Notes

KyaniteLabs/Epoch is an MCP server that provides time estimation capabilities for AI agents. It implements multiple estimation models including PERT, COCOMO II, Monte Carlo simulation, sprint forecasting, and token-to-time mapping for cost and schedule risk analysis.

1 stars on GitHub. Last updated 2026-06-07. Licensed Apache-2.0.

Use cases

• Estimating task durations for AI-driven project planning
• Forecasting sprint timelines and resource allocation
• Mapping token usage to time and cost estimates

Pros

• Supports multiple established estimation models (PERT, COCOMO II, Monte Carlo)
• Integrates with AI agents via the Model Context Protocol
• Written in TypeScript for type safety and broad compatibility

Cons

• Very early stage with only 1 star on GitHub
• Limited community adoption and documentation
• Requires setup and integration effort for custom use

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.