MODAL
Project Intelligence
Your agent has no idea what sprint you're in.
MODAL embeds project management in the IDE and exposes live sprint state, active tickets, and team context to any AI agent via a REST API on localhost.
The Bridge API
A REST API on localhost:7742. Your agent reads project state the same way it reads files — structured data, no auth dance, no context switching. 30 endpoints, SSE streaming, OpenAPI spec included.
Active ticket context
The agent knows what you’re working on before it writes a line of code.
GET /api/v1/context/active-ticket
{
"key": "ENG-1842",
"summary": "Add retry logic to order sync pipeline",
"status": "In Progress",
"assignee": "mauricio",
"acceptance_criteria": [
"Retry up to 3 times with exponential backoff",
"Log each retry attempt to Datadog",
"Alert on-call after final failure"
],
"sprint": "Sprint 24 — ends Apr 11"
}Sprint health
Real-time sprint awareness — not a static report, a live feed.
GET /api/v1/sprint/health
{
"sprint": "Sprint 24",
"progress": "62%",
"days_remaining": 5,
"at_risk": [
{
"key": "ENG-1838",
"reason": "blocked by external API dependency",
"blocked_days": 2
}
],
"velocity_trend": "stable"
}Team collision detection
Know when two people are touching the same code — before the merge conflict.
GET /api/v1/team/active-work
{
"active_branches": 4,
"potential_conflicts": [
{
"files": ["src/services/order_sync.py"],
"branches": ["feat/retry-logic", "fix/sync-timeout"],
"assignees": ["mauricio", "kane"]
}
]
}The context loop
Context flows in, enriches the agent, outputs flow back, state updates. Each cycle makes the next one better.
Watchers
Detect git branch, open files, workspace state
State Store
Live model of tickets, sprint, team activity
Bridge API
HTTP + SSE on localhost for any agent
AI Agent
Receives full context, executes with awareness
Sidebar UI
Renders agent outputs, insights, analysis
Project Tracker
State syncs back to Jira, Linear, GitHub