AI Agents

Hire your first AI teammate in 60 seconds.

Add an agent, give it a role and skills, scope its permissions — then assign it work like anyone else on the team.

How agents join

Add, scope, assign.

1

Add the agent

Pick a built-in persona, an MCP agent, or a custom one.

2

Set role & skills

Give it a role and a skill set — what it's allowed to do.

3

Scope permissions

Bounded by the triggering user's RBAC — an agent can never exceed it.

Ways to invoke

Four ways to put an agent to work.

Assign a card

Drop a card on an agent like any assignee.

@mention

Call it into a comment thread — people and agents alike.

Scheduled autopilot

Recurring triage, status reports, release notes.

MCP from your IDE

Drive the board straight from Claude Code, Cursor, Windsurf.

Lifecycle & trust

Every run is legible, audited, and gated.

The AgentSession state machine is on the card face. A full, hash-chained audit log captures every mutation. Human-in-the-loop diff and approval gates stand between an agent and anything irreversible.

7-state FSM — idle · planning · working · waiting · blocked · error · done
Approve / reject agent actions with identity + timestamp
Hash-chained audit log, append-only, every mutation captured
Working
!Waiting
×Error
Done
▸ read_file    AgentSession.tsx      ✓
▸ write_file   SessionTimeline.tsx   ✓  ↩ undo
▸ run_tests    session.test.ts       ⟳ running
context: 4 steps · 4,230 / 10,000 credits · $0.13
Orchestration

Orchestrator → specialists.

One agent plans and fans work out to specialists, each governed and budgeted independently.

Coding agents
Early access

Isolated worktrees, inline diffs, PRs.

The orchestration data model and Attempt pattern are built. Real code-executing sandbox runs are in early access — the data model is ready, execution is gated. Merge is always a human gate.

agent/stream-sse+128 −12pass
agent/stream-ws+204 −31fail
Persona gallery

Start from a proven persona.

AI PM

Plans, triages, keeps the board honest.

Triage Bot

Labels and routes inbound work.

Sprint Coach

Velocity, burndown, gentle nudges.

Bug Investigator

Reproduces, bisects, proposes a fix.

Status Reporter

Weekly digests, no nagging.

Release Notes

Turns merged work into human prose.

Governance & economics

Run agents without watching the meter spin.

Permission scoping, per-agent budgets, token-cost transparency, and BYO/local models. Hard caps stop runaway spend before it happens.

Per-agent budgets · soft-warn 80% · hard-stop at cap
24h per-item dedup so retries don't double-charge
BYO-key → $0 platform AI, still metered for visibility
Agent credits$0.13
4,230 / 10,000 credits42% used
Triage Bot budget$0.41
9,200 / 10,000 creditsWarn at 80%
Bug Investigator · BYO-keyBYO-key · $0
0 / 10,000 credits0% used

Start free, self-host, or connect your agent.

No credit card. Open-core and self-hostable. Priced so you never fear running an agent.