Agentra LabsAgentra Labs DocsPublic Documentation

Get Started

Integration Guide

How to integrate AgenticTime with other tools, frameworks, and Agentra sisters.

How to integrate AgenticTime with other tools, frameworks, and Agentra sisters.

MCP Integration

AgenticTime runs as an MCP server. Any MCP-compatible client can use it:

  • Claude Desktop
  • Claude Code
  • Cursor
  • Windsurf
  • VS Code (with MCP extension)
  • Codex
  • Any JSON-RPC 2.0 MCP client

Configuration

{
  "mcpServers": {
    "agentic-time": {
      "command": "~/.local/bin/agentic-time-mcp-agentra",
      "args": ["serve"]
    }
  }
}

Sister Integration

AgenticTime + AgenticMemory

Link temporal context to memory nodes:

  • Use decay curves to weight memory freshness when retrieving
  • Attach deadline IDs to decision nodes for temporal context
  • Use time_log to record temporal observations alongside conversation_log

AgenticTime + AgenticVision

Schedule periodic visual captures:

  • Create schedules for regular UI state snapshots
  • Link vision observations to timeline events
  • Track UI changes over time with temporal anchoring

AgenticTime + AgenticCodebase

Track development timelines:

  • Model deployment pipelines as sequences
  • Estimate refactoring durations and track actuals
  • Link code analysis sessions to deadlines

AgenticTime + AgenticIdentity

Audit temporal operations:

  • Sign deadline modifications with identity receipts
  • Track schedule changes with action context
  • Use competence records to improve duration estimates

Python Integration

from agentic_time import TimeGraph

# Open or create
tg = TimeGraph("project.atime")

# Add entities
deadline_id = tg.add_deadline("Ship v1.0", "2026-03-15T17:00:00Z", priority="high")
tg.add_duration_estimate("Auth refactor", hours=8, confidence=0.7)
tg.add_schedule("Daily standup", recurrence="0 9 * * MON-FRI", duration_minutes=15)

# Query
upcoming = tg.list_deadlines(status="pending", sort_by="due_at")
conflicts = tg.check_conflicts("2026-03-10", "2026-03-14")

# Save
tg.save()

REST API (Server Profile)

When running in server mode, AgenticTime accepts JSON-RPC 2.0 requests:

curl -X POST http://localhost:8080 \
  -H "Authorization: Bearer $AGENTIC_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","method":"tools/call","params":{"name":"time_stats"},"id":1}'

Webhook Integration

AgenticTime can be configured to emit webhook notifications for:

  • Deadline approaching (configurable threshold)
  • Deadline missed
  • Schedule conflict detected
  • Sequence step completed

Environment Variables

See Configuration for the full environment variable reference.