Team Management

Teams are first-class meta-agents in Mnemom. Create a team, add agents, derive a collective alignment card, and monitor team reputation — all through the API or dashboard. This guide covers the full team lifecycle: creation, roster management, alignment cards, reputation monitoring, and team badges.

Team operations require the team_reputation feature flag, available on Team and Enterprise plans. See Pricing for details.

Quick start

Create a team, add members, and derive a card in three API calls:

import { createTeam, deriveTeamCard } from '@mnemom/teams';

// 1. Create a team with initial members
const team = await createTeam({
  org_id: 'org-abc123',
  name: 'Support Pipeline Alpha',
  description: 'Customer support agent team',
  agent_ids: ['agent-a', 'agent-b', 'agent-c'],
});

console.log(`Team created: ${team.team.id}`);
console.log(`Members: ${team.members.length}`);

// 2. Derive a team alignment card from member cards
const derived = await deriveTeamCard(team.team.id);
console.log(`Card derived from ${derived.members_with_cards} members`);
console.log(`Card source: ${derived.card_source}`);

import httpx

API_BASE = "https://api.mnemom.ai"
HEADERS = {"Authorization": f"Bearer {api_key}", "Content-Type": "application/json"}

# 1. Create a team with initial members
response = httpx.post(f"{API_BASE}/v1/teams", headers=HEADERS, json={
    "org_id": "org-abc123",
    "name": "Support Pipeline Alpha",
    "description": "Customer support agent team",
    "agent_ids": ["agent-a", "agent-b", "agent-c"],
})
team = response.json()
team_id = team["team"]["id"]
print(f"Team created: {team_id}")

# 2. Derive a team alignment card from member cards
card = httpx.post(f"{API_BASE}/v1/teams/{team_id}/card/derive", headers=HEADERS)
print(f"Card derived from {card.json()['members_with_cards']} members")

Team lifecycle

Create → Add Members → Derive Card → Monitor Reputation → Archive
  │          │              │               │                 │
  ▼          ▼              ▼               ▼                 ▼
POST      POST           POST           GET              DELETE
/teams    /members       /card/derive   /reputation      /teams/{id}

1. Create a team

Teams require an organization, a name, and at least 2 initial members (max 50).

curl -X POST https://api.mnemom.ai/v1/teams \
  -H "Authorization: Bearer $TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "org_id": "org-abc123",
    "name": "Support Pipeline Alpha",
    "description": "Customer support agent team",
    "agent_ids": ["agent-a", "agent-b", "agent-c"],
    "metadata": { "environment": "production", "domain": "support" }
  }'

Team names must be unique within an organization. The response includes the team object and the initial member list.

2. Manage roster

Add or remove members as your team evolves. Every roster change is logged in the audit trail. Add members:

curl -X POST https://api.mnemom.ai/v1/teams/{team_id}/members \
  -H "Authorization: Bearer $TOKEN" \
  -H "Content-Type: application/json" \
  -d '{ "agent_ids": ["agent-d", "agent-e"] }'

The response tells you which agents were added and which were already members (idempotent). Remove a member:

curl -X DELETE https://api.mnemom.ai/v1/teams/{team_id}/members/{agent_id} \
  -H "Authorization: Bearer $TOKEN"

Teams must have at least 2 members. Removing the second-to-last member will fail with a 400 error. Archive the team instead if it’s no longer needed.

View roster history:

curl https://api.mnemom.ai/v1/teams/{team_id}/roster-history \
  -H "Authorization: Bearer $TOKEN"

Returns a paginated audit trail of all roster changes with timestamps, change types, and actors.

3. Team alignment cards

Team alignment cards declare the team’s collective values, autonomy boundaries, and coordination mode. There are two approaches: Auto-derive from members (recommended):

curl -X POST https://api.mnemom.ai/v1/teams/{team_id}/card/derive \
  -H "Authorization: Bearer $TOKEN"

The derivation algorithm merges member cards:

Values are unioned and ordered by frequency
Forbidden actions from any member apply to the team
The strictest audit retention policy wins
Requires at least 2 members with active alignment cards

Set manually:

curl -X PUT https://api.mnemom.ai/v1/teams/{team_id}/card \
  -H "Authorization: Bearer $TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "principal": {
      "type": "organization",
      "identifier": "org-abc123",
      "relationship": "delegated_authority"
    },
    "values": {
      "declared": ["reliability", "transparency", "user_safety"]
    },
    "autonomy_envelope": {
      "bounded_actions": ["respond_to_customer", "escalate_ticket"],
      "forbidden_actions": ["delete_customer_data", "modify_billing"],
      "escalation_triggers": ["customer_complaint", "refund_request_over_100"]
    },
    "audit_commitment": {
      "retention_days": 90
    }
  }'

View card history:

curl https://api.mnemom.ai/v1/teams/{team_id}/card/history \
  -H "Authorization: Bearer $TOKEN"

4. Monitor reputation

Once your team has 10+ risk assessments, a Team Trust Rating is published.

curl -H "Authorization: Bearer $TOKEN" \
  https://api.mnemom.ai/v1/teams/{team_id}/reputation

Track score trends over time:

curl -H "Authorization: Bearer $TOKEN" \
  https://api.mnemom.ai/v1/teams/{team_id}/reputation/history

See Team Trust Rating for details on the 5-component scoring model.

5. Archive a team

Archiving soft-deletes the team. Historical data and reputation scores are preserved.

curl -X DELETE https://api.mnemom.ai/v1/teams/{team_id} \
  -H "Authorization: Bearer $TOKEN"

Team badges

Embed team reputation badges anywhere — READMEs, documentation, dashboards.

GET https://api.mnemom.ai/v1/teams/{team_id}/badge.svg?variant={variant}

Six badge variants are available:

Variant	Display
`score`	`[ Team Trust \| 812 ]`
`grade`	`[ Team Trust \| AA ]`
`score_grade`	`[ Team Trust \| AA 812 ]`
`score_trend`	`[ Team Trust \| AA 812 ↑ ]`
`score_tier`	`[ Mnemom Team \| 812 Established ]`
`compact`	`[ AA ]`

Pre-eligible teams (fewer than 10 assessments) show a progress badge:

[ Team Trust | Building 4/10 ]

Markdown embed:

[![Team Trust Rating](https://api.mnemom.ai/v1/teams/team-abc123/badge.svg?variant=score)](https://www.mnemom.ai/teams/team-abc123/reputation)

HTML embed:

<a href="https://www.mnemom.ai/teams/team-abc123/reputation">
  <img src="https://api.mnemom.ai/v1/teams/team-abc123/badge.svg?variant=score" alt="Team Trust Rating" />
</a>

React:

function TeamBadge({ teamId }) {
  const badgeUrl = `https://api.mnemom.ai/v1/teams/${teamId}/badge.svg?variant=score`;
  const pageUrl = `https://www.mnemom.ai/teams/${teamId}/reputation`;

  return (
    <a href={pageUrl} target="_blank" rel="noopener noreferrer">
      <img src={badgeUrl} alt="Team Trust Rating" />
    </a>
  );
}

Badges are cached for 60 minutes at the CDN edge. For real-time data, use the API directly.

RBAC requirements

Operation	Required Role
Create team	`owner` or `admin`
Update team	`owner` or `admin`
Archive team	`owner` or `admin`
Add/remove members	`owner` or `admin`
Set team card	`owner` or `admin`
Derive team card	`owner` or `admin`
View team details	Any org member
View roster history	Any org member
View card history	Any org member
View reputation	Public (no auth for public teams)

Feature requirements

Team operations require the team_reputation feature flag:

Plan	Team Features
Free	Not available
Developer	Not available
Team	200 team reputation computations included/mo ($0.005/ea overage)
Enterprise	Unlimited

Guides

Safe House

Quick start

Team lifecycle

1. Create a team

2. Manage roster

3. Team alignment cards

4. Monitor reputation

5. Archive a team

Team badges

RBAC requirements

Feature requirements

See also

​Quick start

​Team lifecycle

​1. Create a team

​2. Manage roster

​3. Team alignment cards

​4. Monitor reputation

​5. Archive a team

​Team badges

​RBAC requirements

​Feature requirements

​See also

Quick start

Team lifecycle

1. Create a team

2. Manage roster

3. Team alignment cards

4. Monitor reputation

5. Archive a team

Team badges

RBAC requirements

Feature requirements

See also