Quickstart

Use cases by outcome

Real, end-to-end sequences you can copy and adapt. Start with a flagship; the same calls power use cases across the whole employee lifecycle.

One runbook your agent can run end to end.

Start

Before any call: sign in to create a key, then send it as your bearer token. Key issuance lives in the signed-in portal — the public docs never issue one.

1
Sign in and create a key
Keys are issued in the signed-in portal. Create one and copy it once — you will not see it again.
Request access
2
Keep it server-side
Set the key as an environment variable on your backend. Never ship it to the browser.
Request
```
export THOMAS_KEY="sk_live_…"
```
3
Send it on every call
Pass the key as a bearer token on each request. The base URL (shown on every recipe step and in the API reference — it already includes the version) is your one source of truth; set it once as $THOMAS_API_URL. Each key carries only the scopes it needs — no key grants bulk access to raw scoring or the item bank.
Request
```
curl "$THOMAS_API_URL/company-individuals" \
  -H "Authorization: Bearer $THOMAS_KEY"
```

Build this with your agent

Open with Claude Open with ChatGPT

Open these calls in the API reference

Day one: bring your directory into Thomas and trigger everyone's first assessment, so profiles exist before you build anything on top. About 100 batched calls for 10,000 people, not 10,000.

Validate a batch first (dry run)

Send up to 100 rows to validate. Nothing is written — you get a per-row classification (ready, reactivate or rejected) and a summary, so one bad row never blocks the rest.

POSThttps://api.thomas.co/v1/colleagues/bulk/validate

Request

POST /colleagues/bulk/validate
{ "rows": [{ "email": "alex@acme.com" }] }

Response

{
  "summary": { "ready": 98, "reactivate": 2, "rejected": 0 },
  "results": [{
    "index": 0,
    "email": "alex@acme.com",
    "classification": "ready",
    "archivedProfile": {
      "id": "00000000-0000-4000-8000-000000000081",
      "firstName": "Alex",
      "lastName": "Taylor"
    }
  }]
}

2
Commit the batch
Send the confirmed rows. Idempotent by email: people you already have reactivate instead of duplicating, so re-syncing your directory nightly is safe. Email is the persistent identity key; an HRIS id is scoped to your org. Each created or reactivated row returns its individualId; collect those (rejected rows have none) to assign assessments in the next step.
How identity works
POSThttps://api.thomas.co/v1/colleagues/bulk
Request
```
POST /colleagues/bulk
{ "rows": [{ "email": "alex@acme.com", "classification": "ready" }] }
```
Response
```
{
  "created": 98,
  "reactivated": 2,
  "rejected": 0,
  "details": [{
    "index": 0,
    "email": "alex@acme.com",
    "status": "created",
    "individualId": "00000000-0000-4000-8000-000000000021"
  }]
}
```

Trigger assessments in bulk

Assign the behaviour assessment to up to 100 individuals per call. Already-assessed people are skipped, so it is safe to re-run.

POSThttps://api.thomas.co/v1/assessments/assignments/bulk

Request

POST /assessments/assignments/bulk
{ "individualIds": ["00000000-0000-4000-8000-000000000021"], "assessmentType": "behaviour" }

Response

{
  "created": 1,
  "skipped": 0,
  "results": [{
    "id": "00000000-0000-4000-8000-000000000031",
    "tenantId": "00000000-0000-4000-8000-000000000001",
    "individualId": "00000000-0000-4000-8000-000000000021",
    "assessmentType": "behaviour",
    "assignedBy": "00000000-0000-4000-8000-000000000011",
    "assignedAt": "2026-10-01T00:00:00.000Z",
    "status": "PENDING",
    "createdAt": "2026-10-01T00:00:00.000Z",
    "updatedAt": "2026-10-01T00:00:00.000Z"
  }]
}

Poll an assignment until complete

Poll each assignment until its status is COMPLETED, then read the profile.

GEThttps://api.thomas.co/v1/assessments/assignments/{id}

Request

GET /assessments/assignments/00000000-0000-4000-8000-000000000031

Response

{
  "id": "00000000-0000-4000-8000-000000000031",
  "tenantId": "00000000-0000-4000-8000-000000000001",
  "individualId": "00000000-0000-4000-8000-000000000021",
  "assessmentType": "behaviour",
  "assignedBy": "00000000-0000-4000-8000-000000000011",
  "assignedAt": "2026-10-01T00:00:00.000Z",
  "status": "PENDING",
  "createdAt": "2026-10-01T00:00:00.000Z",
  "updatedAt": "2026-10-01T00:00:00.000Z"
}

List assignments to track progress

List assignments to see how many have completed across the batch.

GEThttps://api.thomas.co/v1/assessments/assignments

Request

GET /assessments/assignments?status=COMPLETED

Response

{
  "data": [{
    "id": "00000000-0000-4000-8000-000000000031",
    "tenantId": "00000000-0000-4000-8000-000000000001",
    "individualId": "00000000-0000-4000-8000-000000000021",
    "assessmentType": "behaviour",
    "assignedBy": "00000000-0000-4000-8000-000000000011",
    "assignedAt": "2026-10-01T00:00:00.000Z",
    "status": "COMPLETED",
    "createdAt": "2026-10-01T00:00:00.000Z",
    "updatedAt": "2026-10-02T09:30:00.000Z"
  }],
  "total": 87
}

6
React to completion (webhook)
Coming soon
A result webhook (an assessment.completed event) is coming soon. Until then, poll the assignment status as above.

Build this with your agent

Open with Claude Open with ChatGPT

Open these calls in the API reference

People & profiles

Trigger a Thomas assessment for one person (a candidate who applies, or an employee), then read their behaviour profile.

1
Resolve the person
Resolve a candidate or employee to one Thomas identity by email — the persistent, case-insensitive identity key — or by your org-scoped HRIS id. Prefer not to send an email? Resolve by your own opaque partner reference instead.
How identity works
POSThttps://api.thomas.co/v1/individuals/resolve
Request
```
POST /individuals/resolve
{ "email": "alex@acme.com" }
```
Response
```
{
  "individualId": "00000000-0000-4000-8000-000000000021",
  "created": true
}
```
2
Enrol as a candidate
Resolve returns a global identity with no tenant membership, so an org-scoped call would reject it. Enrol the resolved person into your tenant as a candidate — relationshipType candidate, never employee — before you assign the assessment.
How identity works
POSThttps://api.thomas.co/v1/company-individuals
Request
```
POST /company-individuals
{ "individualId": "00000000-0000-4000-8000-000000000021", "role": "viewer", "relationshipType": "candidate" }
```

Assign the assessment

Assign the behaviour assessment to the resolved individual.

POSThttps://api.thomas.co/v1/assessments/assignments

Request

POST /assessments/assignments
{ "individualId": "00000000-0000-4000-8000-000000000021", "assessmentType": "behaviour" }

Response

{
  "id": "00000000-0000-4000-8000-000000000031",
  "tenantId": "00000000-0000-4000-8000-000000000001",
  "individualId": "00000000-0000-4000-8000-000000000021",
  "assessmentType": "behaviour",
  "assignedBy": "00000000-0000-4000-8000-000000000011",
  "assignedAt": "2026-10-01T00:00:00.000Z",
  "status": "PENDING",
  "createdAt": "2026-10-01T00:00:00.000Z",
  "updatedAt": "2026-10-01T00:00:00.000Z"
}

4
The person completes the assessment
Now the person takes the assessment. They complete and submit it themselves in Thomas — the API cannot start, complete, or score it on their behalf. Their behaviour profile becomes available only once they have finished and it has been scored; until then the profile read below returns 404 ("No scored assessment found for this individual").

Poll the assignment

Poll the assignment until its status is COMPLETED. That status flips only once the person has submitted the assessment (the previous step) — not from the assignment being created.

GEThttps://api.thomas.co/v1/assessments/assignments/{id}

Request

GET /assessments/assignments/00000000-0000-4000-8000-000000000031

Response

{
  "id": "00000000-0000-4000-8000-000000000031",
  "tenantId": "00000000-0000-4000-8000-000000000001",
  "individualId": "00000000-0000-4000-8000-000000000021",
  "assessmentType": "behaviour",
  "assignedBy": "00000000-0000-4000-8000-000000000011",
  "assignedAt": "2026-10-01T00:00:00.000Z",
  "status": "COMPLETED",
  "resultId": "00000000-0000-4000-8000-000000000041",
  "createdAt": "2026-10-01T00:00:00.000Z",
  "updatedAt": "2026-10-02T09:30:00.000Z"
}

Read the behaviour profile

Once the person has completed the assessment and it has been scored, fetch their validated behaviour profile. Before a scored result exists this returns 404 ("No scored assessment found for this individual").

GEThttps://api.thomas.co/v1/assessments/profile/{individualId}

Request

GET /assessments/profile/00000000-0000-4000-8000-000000000021

Response

{
  "connectionPowerType": "action",
  "traitSpectrums": [
    { "dimension": "D", "name": "Dominance", "leftPole": "Accommodating", "rightPole": "Direct", "position": 82, "band": "High" },
    { "dimension": "I", "name": "Influence", "leftPole": "Reflective", "rightPole": "Outgoing", "position": 61, "band": "High" },
    { "dimension": "S", "name": "Steadiness", "leftPole": "Spontaneous", "rightPole": "Methodical", "position": 34, "band": "Low" },
    { "dimension": "C", "name": "Compliance", "leftPole": "Pragmatist", "rightPole": "Perfectionist", "position": 47, "band": "Low" }
  ],
  "selfImageScores": { "d": 78, "i": 55, "s": 30, "c": 41 },
  "discAdvanced": { "workMaskScores": { "d": 71, "i": 60, "s": 28, "c": 44 }, "underPressureScores": { "d": 66, "i": 52, "s": 35, "c": 49 } },
  "scoredAt": "2026-10-02T09:30:00.000Z",
  "soloInsights": [{
    "topic": {
      "id": "00000000-0000-4000-8000-000000000051",
      "assessmentType": "behaviour",
      "slug": "communication-style",
      "label": "Communication style",
      "sortOrder": 1,
      "createdAt": "2026-09-01T00:00:00.000Z",
      "updatedAt": "2026-09-01T00:00:00.000Z"
    },
    "entries": [{
      "id": "00000000-0000-4000-8000-000000000061",
      "topicId": "00000000-0000-4000-8000-000000000051",
      "contentType": "markdown",
      "mappingKey": "action-profile",
      "mappingGranularity": "profile",
      "title": "Lead with action",
      "body": "Start with the outcome, then agree clear next steps.",
      "sortOrder": 1,
      "createdAt": "2026-09-01T00:00:00.000Z",
      "updatedAt": "2026-09-01T00:00:00.000Z"
    }]
  }]
}

7
React to completion (webhook)
Coming soon
Result webhooks (an assessment.completed event) are coming soon. For now, poll the assignment status.

Build this with your agent

Open with Claude Open with ChatGPT

Open these calls in the API reference

Pull a person's profile into your review screen, so a manager sees how to get the best from them, not just a rating.

1
Resolve the person
Resolve the reviewee to one Thomas identity by email — the persistent identity key — or by your org-scoped HRIS id.
How identity works
POSThttps://api.thomas.co/v1/individuals/resolve
Request
```
POST /individuals/resolve
{ "email": "alex@acme.com" }
```
Response
```
{
  "individualId": "00000000-0000-4000-8000-000000000021",
  "created": true
}
```
2
Enrol as a colleague
Resolve returns a global identity with no tenant membership, so the profile read would reject it. Enrol the reviewee as a colleague — createColleague resolves and enrols by email in one call, as relationshipType employee — before you read their profile.
How identity works
POSThttps://api.thomas.co/v1/colleagues
Request
```
POST /colleagues
{ "email": "alex@acme.com" }
```

Read the behaviour profile

Fetch the trait spectrums and the "how to work with this person" insight. This reads only once the person has completed a Thomas behaviour assessment — something they do themselves; it cannot be triggered on their behalf through the API, and until then it returns 404 ("No scored assessment found for this individual").

GEThttps://api.thomas.co/v1/assessments/profile/{individualId}

Request

GET /assessments/profile/00000000-0000-4000-8000-000000000021

Response

{
  "connectionPowerType": "energy",
  "traitSpectrums": [
    { "dimension": "D", "name": "Dominance", "leftPole": "Accommodating", "rightPole": "Direct", "position": 58, "band": "High" },
    { "dimension": "I", "name": "Influence", "leftPole": "Reflective", "rightPole": "Outgoing", "position": 64, "band": "High" },
    { "dimension": "S", "name": "Steadiness", "leftPole": "Spontaneous", "rightPole": "Methodical", "position": 41, "band": "Low" },
    { "dimension": "C", "name": "Compliance", "leftPole": "Pragmatist", "rightPole": "Perfectionist", "position": 38, "band": "Low" }
  ],
  "selfImageScores": { "d": 62, "i": 70, "s": 44, "c": 38 },
  "discAdvanced": { "workMaskScores": { "d": 58, "i": 66, "s": 49, "c": 41 }, "underPressureScores": { "d": 64, "i": 61, "s": 40, "c": 45 } },
  "scoredAt": "2026-10-02T09:30:00.000Z",
  "soloInsights": [{
    "topic": {
      "id": "00000000-0000-4000-8000-000000000052",
      "assessmentType": "behaviour",
      "slug": "collaboration-style",
      "label": "Collaboration style",
      "sortOrder": 1,
      "createdAt": "2026-09-01T00:00:00.000Z",
      "updatedAt": "2026-09-01T00:00:00.000Z"
    },
    "entries": [{
      "id": "00000000-0000-4000-8000-000000000062",
      "topicId": "00000000-0000-4000-8000-000000000052",
      "contentType": "markdown",
      "mappingKey": "energy-profile",
      "mappingGranularity": "profile",
      "title": "Keep the conversation lively",
      "body": "Use discussion and visible momentum to build alignment.",
      "sortOrder": 1,
      "createdAt": "2026-09-01T00:00:00.000Z",
      "updatedAt": "2026-09-01T00:00:00.000Z"
    }]
  }]
}

4
Render the insight in your review UIyour system
Show the trait spectrums and the "how to work with this person" insight alongside the manager's ratings — in your own review screen.

Build this with your agent

Open with Claude Open with ChatGPT

Open these calls in the API reference

Surface how to communicate with a specific person, right where the conversation already happens.

1
Resolve the person
Resolve the person to one Thomas identity by email — the persistent identity key — or by your org-scoped HRIS id.
How identity works
POSThttps://api.thomas.co/v1/individuals/resolve
Request
```
POST /individuals/resolve
{ "email": "alex@acme.com" }
```
Response
```
{
  "individualId": "00000000-0000-4000-8000-000000000021",
  "created": true
}
```
2
Enrol as a colleague
Resolve returns a global identity with no tenant membership, so the profile read would reject it. Enrol the person as a colleague — createColleague resolves and enrols by email in one call, as relationshipType employee — before you read their profile.
How identity works
POSThttps://api.thomas.co/v1/colleagues
Request
```
POST /colleagues
{ "email": "alex@acme.com" }
```

Read the behaviour profile

GEThttps://api.thomas.co/v1/assessments/profile/{individualId}

Request

GET /assessments/profile/00000000-0000-4000-8000-000000000021

Response

{
  "connectionPowerType": "energy",
  "traitSpectrums": [
    { "dimension": "D", "name": "Dominance", "leftPole": "Accommodating", "rightPole": "Direct", "position": 58, "band": "High" },
    { "dimension": "I", "name": "Influence", "leftPole": "Reflective", "rightPole": "Outgoing", "position": 64, "band": "High" },
    { "dimension": "S", "name": "Steadiness", "leftPole": "Spontaneous", "rightPole": "Methodical", "position": 41, "band": "Low" },
    { "dimension": "C", "name": "Compliance", "leftPole": "Pragmatist", "rightPole": "Perfectionist", "position": 38, "band": "Low" }
  ],
  "selfImageScores": { "d": 62, "i": 70, "s": 44, "c": 38 },
  "discAdvanced": { "workMaskScores": { "d": 58, "i": 66, "s": 49, "c": 41 }, "underPressureScores": { "d": 64, "i": 61, "s": 40, "c": 45 } },
  "scoredAt": "2026-10-02T09:30:00.000Z",
  "soloInsights": [{
    "topic": {
      "id": "00000000-0000-4000-8000-000000000052",
      "assessmentType": "behaviour",
      "slug": "collaboration-style",
      "label": "Collaboration style",
      "sortOrder": 1,
      "createdAt": "2026-09-01T00:00:00.000Z",
      "updatedAt": "2026-09-01T00:00:00.000Z"
    },
    "entries": [{
      "id": "00000000-0000-4000-8000-000000000062",
      "topicId": "00000000-0000-4000-8000-000000000052",
      "contentType": "markdown",
      "mappingKey": "energy-profile",
      "mappingGranularity": "profile",
      "title": "Keep the conversation lively",
      "body": "Use discussion and visible momentum to build alignment.",
      "sortOrder": 1,
      "createdAt": "2026-09-01T00:00:00.000Z",
      "updatedAt": "2026-09-01T00:00:00.000Z"
    }]
  }]
}

4
Nudge in your messaging platformyour system
Surface the "how to work with this person" insight in your messaging platform at the right moment.

Build this with your agent

Open with Claude Open with ChatGPT

Open these calls in the API reference

Recommend the next learning step from a person's profile.

1
Resolve the person
Resolve the person to one Thomas identity by email — the persistent identity key — or by your org-scoped HRIS id.
How identity works
POSThttps://api.thomas.co/v1/individuals/resolve
Request
```
POST /individuals/resolve
{ "email": "alex@acme.com" }
```
Response
```
{
  "individualId": "00000000-0000-4000-8000-000000000021",
  "created": true
}
```
2
Enrol as a colleague
Resolve returns a global identity with no tenant membership, so the profile read would reject it. Enrol the person as a colleague — createColleague resolves and enrols by email in one call, as relationshipType employee — before you read their profile.
How identity works
POSThttps://api.thomas.co/v1/colleagues
Request
```
POST /colleagues
{ "email": "alex@acme.com" }
```

Read the behaviour profile

Fetch the trait spectrums that inform the recommendation. This reads only once the person has completed a Thomas behaviour assessment — something they do themselves; it cannot be triggered on their behalf through the API, and until then it returns 404 ("No scored assessment found for this individual").

GEThttps://api.thomas.co/v1/assessments/profile/{individualId}

Request

GET /assessments/profile/00000000-0000-4000-8000-000000000021

Response

{
  "connectionPowerType": "energy",
  "traitSpectrums": [
    { "dimension": "D", "name": "Dominance", "leftPole": "Accommodating", "rightPole": "Direct", "position": 58, "band": "High" },
    { "dimension": "I", "name": "Influence", "leftPole": "Reflective", "rightPole": "Outgoing", "position": 64, "band": "High" },
    { "dimension": "S", "name": "Steadiness", "leftPole": "Spontaneous", "rightPole": "Methodical", "position": 41, "band": "Low" },
    { "dimension": "C", "name": "Compliance", "leftPole": "Pragmatist", "rightPole": "Perfectionist", "position": 38, "band": "Low" }
  ],
  "selfImageScores": { "d": 62, "i": 70, "s": 44, "c": 38 },
  "discAdvanced": { "workMaskScores": { "d": 58, "i": 66, "s": 49, "c": 41 }, "underPressureScores": { "d": 64, "i": 61, "s": 40, "c": 45 } },
  "scoredAt": "2026-10-02T09:30:00.000Z",
  "soloInsights": [{
    "topic": {
      "id": "00000000-0000-4000-8000-000000000052",
      "assessmentType": "behaviour",
      "slug": "collaboration-style",
      "label": "Collaboration style",
      "sortOrder": 1,
      "createdAt": "2026-09-01T00:00:00.000Z",
      "updatedAt": "2026-09-01T00:00:00.000Z"
    },
    "entries": [{
      "id": "00000000-0000-4000-8000-000000000062",
      "topicId": "00000000-0000-4000-8000-000000000052",
      "contentType": "markdown",
      "mappingKey": "energy-profile",
      "mappingGranularity": "profile",
      "title": "Keep the conversation lively",
      "body": "Use discussion and visible momentum to build alignment.",
      "sortOrder": 1,
      "createdAt": "2026-09-01T00:00:00.000Z",
      "updatedAt": "2026-09-01T00:00:00.000Z"
    }]
  }]
}

4
Map to learning pathsyour system
Map the profile to paths in your LMS: a doer gets a short video plus a hands-on exercise, not long reading.

Build this with your agent

Open with Claude Open with ChatGPT

Open these calls in the API reference

Give a manager grounded guidance on each report, on demand.

1
Resolve the report
Resolve the report to one Thomas identity by email — the persistent identity key — or by your org-scoped HRIS id.
How identity works
POSThttps://api.thomas.co/v1/individuals/resolve
Request
```
POST /individuals/resolve
{ "email": "alex@acme.com" }
```
Response
```
{
  "individualId": "00000000-0000-4000-8000-000000000021",
  "created": true
}
```
2
Enrol as a colleague
Resolve returns a global identity with no tenant membership, so the profile read would reject it. Enrol the report as a colleague — createColleague resolves and enrols by email in one call, as relationshipType employee — before you read their profile.
How identity works
POSThttps://api.thomas.co/v1/colleagues
Request
```
POST /colleagues
{ "email": "alex@acme.com" }
```

Read the report's profile

Fetch the trait spectrums and self-image scores that ground the manager's coaching. This reads only once the person has completed a Thomas behaviour assessment — something they do themselves; it cannot be triggered on their behalf through the API, and until then it returns 404 ("No scored assessment found for this individual").

GEThttps://api.thomas.co/v1/assessments/profile/{individualId}

Request

GET /assessments/profile/00000000-0000-4000-8000-000000000021

Response

{
  "connectionPowerType": "energy",
  "traitSpectrums": [
    { "dimension": "D", "name": "Dominance", "leftPole": "Accommodating", "rightPole": "Direct", "position": 58, "band": "High" },
    { "dimension": "I", "name": "Influence", "leftPole": "Reflective", "rightPole": "Outgoing", "position": 64, "band": "High" },
    { "dimension": "S", "name": "Steadiness", "leftPole": "Spontaneous", "rightPole": "Methodical", "position": 41, "band": "Low" },
    { "dimension": "C", "name": "Compliance", "leftPole": "Pragmatist", "rightPole": "Perfectionist", "position": 38, "band": "Low" }
  ],
  "selfImageScores": { "d": 62, "i": 70, "s": 44, "c": 38 },
  "discAdvanced": { "workMaskScores": { "d": 58, "i": 66, "s": 49, "c": 41 }, "underPressureScores": { "d": 64, "i": 61, "s": 40, "c": 45 } },
  "scoredAt": "2026-10-02T09:30:00.000Z",
  "soloInsights": [{
    "topic": {
      "id": "00000000-0000-4000-8000-000000000052",
      "assessmentType": "behaviour",
      "slug": "collaboration-style",
      "label": "Collaboration style",
      "sortOrder": 1,
      "createdAt": "2026-09-01T00:00:00.000Z",
      "updatedAt": "2026-09-01T00:00:00.000Z"
    },
    "entries": [{
      "id": "00000000-0000-4000-8000-000000000062",
      "topicId": "00000000-0000-4000-8000-000000000052",
      "contentType": "markdown",
      "mappingKey": "energy-profile",
      "mappingGranularity": "profile",
      "title": "Keep the conversation lively",
      "body": "Use discussion and visible momentum to build alignment.",
      "sortOrder": 1,
      "createdAt": "2026-09-01T00:00:00.000Z",
      "updatedAt": "2026-09-01T00:00:00.000Z"
    }]
  }]
}

Open a manager-context conversation

Ask for coaching about the report; the reply streams back (SSE).

POSThttps://api.thomas.co/v1/ai-coaching/conversations

Request

POST /ai-coaching/conversations
{ "contextTag": "manager", "entryPoint": "prompt_card" }

Response

{
  "id": "00000000-0000-4000-8000-000000000091",
  "contextTag": "manager",
  "entryPoint": "prompt_card",
  "pageId": "individual",
  "title": "Coaching conversation",
  "lastMessageAt": "2026-10-02T09:30:00.000Z",
  "createdAt": "2026-10-02T09:30:00.000Z",
  "updatedAt": "2026-10-02T09:30:00.000Z"
}

Build this with your agent

Open with Claude Open with ChatGPT

Open these calls in the API reference

Ask Thomas for grounded coaching, in the flow of work. Two calls, and the reply streams back as Server-Sent Events.

Open a coaching conversation

Create a coaching conversation with a context: personal, manager, or org analyst.

POSThttps://api.thomas.co/v1/ai-coaching/conversations

Request

POST /ai-coaching/conversations
{ "contextTag": "personal", "entryPoint": "prompt_card" }

Response

{
  "id": "00000000-0000-4000-8000-000000000091",
  "contextTag": "personal",
  "entryPoint": "prompt_card",
  "pageId": "individual",
  "title": "Coaching conversation",
  "lastMessageAt": "2026-10-02T09:30:00.000Z",
  "createdAt": "2026-10-02T09:30:00.000Z",
  "updatedAt": "2026-10-02T09:30:00.000Z"
}

Send a message, stream the reply

POST your intent to the conversation. The answer streams back as Server-Sent Events (tokens), not a single JSON body — render it as it arrives.

POSThttps://api.thomas.co/v1/ai-coaching/conversations/{conversationId}/messages

Request

POST /ai-coaching/conversations/00000000-0000-4000-8000-000000000091/messages
{ "intent": "How should I prepare for my own performance review?", "clientMessageId": "00000000-0000-4000-8000-000000000031", "pageId": "individual", "pageParams": { "individualId": "00000000-0000-4000-8000-000000000022" }, "entryPoint": "prompt_card" }

Response

event: token
data: {"kind":"token","text":"Start by naming what you want out of the review..."}

event: done
data: {"kind":"done"}

3
Surface it where work happensyour system
Drop the streamed coaching into your review screen, messaging platform, or wherever work already happens.

Build this with your agent

Open with Claude Open with ChatGPT

Open these calls in the API reference

Teams & connection

Group resolved people into a team and read its dynamic: the spread of connection powers and where to watch out. No single team score here — that is the tracking recipe.

1
Enrol the team members
Only enrolled colleagues can be grouped into a team — a resolved-but-unenrolled member is rejected. Enrol every member you will reference in the team in one bulk call (resolves and enrols by email, idempotent) as relationshipType employee: send one row per member, then reference each enrolled individual by individualId when you create the team below.
How identity works
POSThttps://api.thomas.co/v1/colleagues/bulk
Request
```
POST /colleagues/bulk
{ "rows": [{ "email": "alex@acme.com", "classification": "ready" }, { "email": "jordan@acme.com", "classification": "ready" }] }
```
Response
```
{
  "created": 2,
  "reactivated": 0,
  "rejected": 0,
  "details": [{
    "index": 0,
    "email": "alex@acme.com",
    "status": "created"
  }, {
    "index": 1,
    "email": "jordan@acme.com",
    "status": "created"
  }]
}
```

Create the team

Group the resolved individuals into a team.

POSThttps://api.thomas.co/v1/teams

Request

POST /teams
{ "name": "Revenue", "memberIndividualIds": ["00000000-0000-4000-8000-000000000021", "00000000-0000-4000-8000-000000000022"] }

Response

{
  "id": "00000000-0000-4000-8000-000000000041",
  "tenantId": "00000000-0000-4000-8000-000000000001",
  "name": "Revenue",
  "teamType": "explicit",
  "sourceManagerIndividualId": null,
  "visibility": "visible",
  "firstSharedAt": null,
  "needsOwnerAssignment": false,
  "deletedAt": null,
  "inactiveAt": null,
  "createdAt": "2026-07-02T10:00:00.000Z",
  "updatedAt": "2026-07-02T10:00:00.000Z",
  "members": [{
    "id": "00000000-0000-4000-8000-000000000061",
    "teamId": "00000000-0000-4000-8000-000000000041",
    "individualId": "00000000-0000-4000-8000-000000000021",
    "companyIndividualId": "00000000-0000-4000-8000-000000000031",
    "tenantId": "00000000-0000-4000-8000-000000000001",
    "role": "member",
    "status": "active",
    "createdAt": "2026-07-02T10:00:00.000Z",
    "updatedAt": "2026-07-02T10:00:00.000Z"
  }]
}

Read the dynamic

Get the power distribution, the dominant power, and strengths/watch-outs. Privacy-safe: a minimum group of three.

GEThttps://api.thomas.co/v1/teams/{teamId}/dynamics

Request

GET /teams/00000000-0000-4000-8000-000000000041/dynamics

Response

{
  "teamId": "00000000-0000-4000-8000-000000000041",
  "members": [{
    "individualId": "00000000-0000-4000-8000-000000000021",
    "displayName": "Alex",
    "connectionPower": "action",
    "behaviourScales": {
      "accommodatingDirect": 72,
      "reflectiveOutgoing": 45,
      "spontaneousMethodical": 88,
      "pragmaticPerfectionist": 33
    },
    "assessmentStatus": "assessed"
  }],
  "powerDistribution": { "action": 1, "energy": 0, "stability": 0, "precision": 0 },
  "dominantPower": "action",
  "narrativeContent": null,
  "strengthsAndWatchouts": [{
    "type": "strength",
    "title": "Fast execution",
    "description": "The team moves quickly once aligned.",
    "narrativeContent": null
  }],
  "scaleDescriptions": null
}

Build this with your agent

Open with Claude Open with ChatGPT

Open these calls in the API reference

Read a team's connection score and how it is trending across runs.

1
Enrol the team members
Only enrolled colleagues can be grouped into a team — a resolved-but-unenrolled member is rejected. Enrol every member you will reference in the team in one bulk call (resolves and enrols by email, idempotent) as relationshipType employee: send one row per member, then reference each enrolled individual by individualId when you create the team below.
How identity works
POSThttps://api.thomas.co/v1/colleagues/bulk
Request
```
POST /colleagues/bulk
{ "rows": [{ "email": "alex@acme.com", "classification": "ready" }, { "email": "jordan@acme.com", "classification": "ready" }] }
```
Response
```
{
  "created": 2,
  "reactivated": 0,
  "rejected": 0,
  "details": [{
    "index": 0,
    "email": "alex@acme.com",
    "status": "created"
  }, {
    "index": 1,
    "email": "jordan@acme.com",
    "status": "created"
  }]
}
```

Create or reuse the team

Group the individuals whose connection you want to track over time.

POSThttps://api.thomas.co/v1/teams

Request

POST /teams
{ "name": "Revenue", "memberIndividualIds": ["00000000-0000-4000-8000-000000000021", "00000000-0000-4000-8000-000000000022"] }

Response

{
  "id": "00000000-0000-4000-8000-000000000051",
  "tenantId": "00000000-0000-4000-8000-000000000001",
  "name": "Revenue",
  "teamType": "explicit",
  "sourceManagerIndividualId": null,
  "visibility": "visible",
  "firstSharedAt": null,
  "needsOwnerAssignment": false,
  "deletedAt": null,
  "inactiveAt": null,
  "createdAt": "2026-07-02T10:00:00.000Z",
  "updatedAt": "2026-07-02T10:00:00.000Z",
  "members": [{
    "id": "00000000-0000-4000-8000-000000000071",
    "teamId": "00000000-0000-4000-8000-000000000051",
    "individualId": "00000000-0000-4000-8000-000000000021",
    "companyIndividualId": "00000000-0000-4000-8000-000000000031",
    "tenantId": "00000000-0000-4000-8000-000000000001",
    "role": "member",
    "status": "active",
    "createdAt": "2026-07-02T10:00:00.000Z",
    "updatedAt": "2026-07-02T10:00:00.000Z"
  }]
}

Read the team results page

Get the team results page: the aggregate score, its six-dimension breakdown, and the trend versus prior runs.

GEThttps://api.thomas.co/v1/teams/{teamId}/runs/{runId}/team-results-page

Request

GET /teams/00000000-0000-4000-8000-000000000051/runs/00000000-0000-4000-8000-000000000052/team-results-page

Response

{
  "teamResult": { "runId": "00000000-0000-4000-8000-000000000052", "teamId": "00000000-0000-4000-8000-000000000051", "dimensions": [{ "dimension": "appreciation", "score": 78, "banding": "MODERATE" }, { "dimension": "belonging", "score": 74, "banding": "MODERATE" }, { "dimension": "cohesion", "score": 64, "banding": "LOW" }, { "dimension": "contribution", "score": 76, "banding": "MODERATE" }, { "dimension": "trust", "score": 82, "banding": "MODERATE" }, { "dimension": "wellbeing", "score": 70, "banding": "MODERATE" }], "overallScore": 72, "overallBanding": "MODERATE", "participantCount": 8, "scoresAvailable": true, "participantsNeeded": 0, "trendDeltas": null, "overallTrend": null, "focusArea": "cohesion" },
  "teamName": "Revenue",
  "teamId": "00000000-0000-4000-8000-000000000051",
  "teamType": "explicit",
  "runPeriod": "2026-07",
  "closedAt": "2026-07-02T10:15:00Z",
  "availablePeriods": ["2026-07", "2026-04"],
  "previousResult": null,
  "narratives": null,
  "trailingSeries": { "overall": [68, 72], "dimensions": { "appreciation": [74, 78], "belonging": [70, 74], "cohesion": [60, 64], "contribution": [72, 76], "trust": [80, 82], "wellbeing": [68, 70] } },
  "viewerHasCompleted": true
}

Schedule recurring runs (optional)

Optionally configure a recurring cadence so the trend keeps updating on its own.

PUThttps://api.thomas.co/v1/organisation/teams/{teamId}/recurrence

Request

PUT /organisation/teams/00000000-0000-4000-8000-000000000051/recurrence
{ "interval": "quarterly", "nextRunDate": "2026-10-01T00:00:00.000Z", "enabled": true }

Response

{
  "interval": "quarterly",
  "nextRunDate": "2026-10-01T00:00:00.000Z",
  "enabled": true,
  "measureId": "00000000-0000-4000-8000-000000000071",
  "teamId": "00000000-0000-4000-8000-000000000051",
  "recurrencePersisted": true
}

Build this with your agent

Open with Claude Open with ChatGPT

Open these calls in the API reference

Compose or rebalance a team using its connection dynamic.

1
Enrol the team members
Only enrolled colleagues can be grouped into a team — a resolved-but-unenrolled member is rejected. Enrol every member you will reference in the team in one bulk call (resolves and enrols by email, idempotent) as relationshipType employee: send one row per member, then reference each enrolled individual by individualId when you create the team below.
How identity works
POSThttps://api.thomas.co/v1/colleagues/bulk
Request
```
POST /colleagues/bulk
{ "rows": [{ "email": "alex@acme.com", "classification": "ready" }, { "email": "jordan@acme.com", "classification": "ready" }] }
```
Response
```
{
  "created": 2,
  "reactivated": 0,
  "rejected": 0,
  "details": [{
    "index": 0,
    "email": "alex@acme.com",
    "status": "created"
  }, {
    "index": 1,
    "email": "jordan@acme.com",
    "status": "created"
  }]
}
```

Create the team

Group the people you are considering into a team.

POSThttps://api.thomas.co/v1/teams

Request

POST /teams
{ "name": "Revenue", "memberIndividualIds": ["00000000-0000-4000-8000-000000000021", "00000000-0000-4000-8000-000000000022"] }

Response

{
  "id": "00000000-0000-4000-8000-000000000041",
  "tenantId": "00000000-0000-4000-8000-000000000001",
  "name": "Revenue",
  "teamType": "explicit",
  "sourceManagerIndividualId": null,
  "visibility": "visible",
  "firstSharedAt": null,
  "needsOwnerAssignment": false,
  "deletedAt": null,
  "inactiveAt": null,
  "createdAt": "2026-07-02T10:00:00.000Z",
  "updatedAt": "2026-07-02T10:00:00.000Z",
  "members": [{
    "id": "00000000-0000-4000-8000-000000000081",
    "teamId": "00000000-0000-4000-8000-000000000041",
    "individualId": "00000000-0000-4000-8000-000000000021",
    "companyIndividualId": "00000000-0000-4000-8000-000000000031",
    "tenantId": "00000000-0000-4000-8000-000000000001",
    "role": "member",
    "status": "active",
    "createdAt": "2026-07-02T10:00:00.000Z",
    "updatedAt": "2026-07-02T10:00:00.000Z"
  }]
}

Read the balance

See the power spread and where to add or rebalance. Privacy-safe: a minimum group of three.

GEThttps://api.thomas.co/v1/teams/{teamId}/dynamics

Request

GET /teams/00000000-0000-4000-8000-000000000041/dynamics

Response

{
  "teamId": "00000000-0000-4000-8000-000000000041",
  "members": [{
    "individualId": "00000000-0000-4000-8000-000000000021",
    "displayName": "Alex",
    "connectionPower": "action",
    "behaviourScales": {
      "accommodatingDirect": 72,
      "reflectiveOutgoing": 45,
      "spontaneousMethodical": 88,
      "pragmaticPerfectionist": 33
    },
    "assessmentStatus": "assessed"
  }],
  "powerDistribution": { "action": 1, "energy": 0, "stability": 0, "precision": 0 },
  "dominantPower": "action",
  "narrativeContent": null,
  "strengthsAndWatchouts": [{
    "type": "strength",
    "title": "Fast execution",
    "description": "The team moves quickly once aligned.",
    "narrativeContent": null
  }],
  "scaleDescriptions": null
}

Build this with your agent

Open with Claude Open with ChatGPT

Open these calls in the API reference

Lifecycle

Weave Thomas across the systems a new joiner touches, personalised to how they work. The one multi-system example: Thomas calls sit alongside your own HRIS, LMS and calendar.

1
Create the joiner in your HRISyour system
Your HRIS creates the new-hire record — the source event that kicks everything off.
2
Enrol the joiner as a colleague
Resolve returns a global identity with no tenant membership, so day-one org-scoped calls would reject the joiner. Enrol them as a colleague — createColleague resolves and enrols by email in one call, as relationshipType employee — as soon as their record exists. Email is the persistent identity key.
How identity works
POSThttps://api.thomas.co/v1/colleagues
Request
```
POST /colleagues
{ "email": "jimmy.page@acme.com" }
```

Assign the assessment on day one

Assign the behaviour assessment straight away, so a profile exists before their first week.

POSThttps://api.thomas.co/v1/assessments/assignments

Request

POST /assessments/assignments
{ "individualId": "00000000-0000-4000-8000-000000000022", "assessmentType": "behaviour" }

Response

{
  "id": "00000000-0000-4000-8000-000000000032",
  "tenantId": "00000000-0000-4000-8000-000000000001",
  "individualId": "00000000-0000-4000-8000-000000000022",
  "assessmentType": "behaviour",
  "assignedBy": "00000000-0000-4000-8000-000000000011",
  "assignedAt": "2026-10-01T00:00:00.000Z",
  "status": "PENDING",
  "createdAt": "2026-10-01T00:00:00.000Z",
  "updatedAt": "2026-10-01T00:00:00.000Z"
}

Read their profile

Once completed, pull the profile to personalise everything downstream. This reads only once the person has completed a Thomas behaviour assessment — something they do themselves; it cannot be triggered on their behalf through the API, and until then it returns 404 ("No scored assessment found for this individual").

GEThttps://api.thomas.co/v1/assessments/profile/{individualId}

Request

GET /assessments/profile/00000000-0000-4000-8000-000000000022

Response

{
  "connectionPowerType": "action",
  "traitSpectrums": [
    { "dimension": "D", "name": "Dominance", "leftPole": "Accommodating", "rightPole": "Direct", "position": 71, "band": "High" },
    { "dimension": "I", "name": "Influence", "leftPole": "Reflective", "rightPole": "Outgoing", "position": 68, "band": "High" },
    { "dimension": "S", "name": "Steadiness", "leftPole": "Spontaneous", "rightPole": "Methodical", "position": 39, "band": "Low" },
    { "dimension": "C", "name": "Compliance", "leftPole": "Pragmatist", "rightPole": "Perfectionist", "position": 43, "band": "Low" }
  ],
  "selfImageScores": { "d": 74, "i": 58, "s": 36, "c": 40 },
  "discAdvanced": { "workMaskScores": { "d": 67, "i": 63, "s": 34, "c": 43 }, "underPressureScores": { "d": 78, "i": 52, "s": 41, "c": 48 } },
  "scoredAt": "2026-10-02T09:30:00.000Z",
  "soloInsights": [{
    "topic": {
      "id": "00000000-0000-4000-8000-000000000053",
      "assessmentType": "behaviour",
      "slug": "onboarding-style",
      "label": "Onboarding style",
      "sortOrder": 1,
      "createdAt": "2026-09-01T00:00:00.000Z",
      "updatedAt": "2026-09-01T00:00:00.000Z"
    },
    "entries": [{
      "id": "00000000-0000-4000-8000-000000000063",
      "topicId": "00000000-0000-4000-8000-000000000053",
      "contentType": "markdown",
      "mappingKey": "action-onboarding",
      "mappingGranularity": "profile",
      "title": "Start with quick wins",
      "body": "Give this joiner visible progress in their first week.",
      "sortOrder": 1,
      "createdAt": "2026-09-01T00:00:00.000Z",
      "updatedAt": "2026-09-01T00:00:00.000Z"
    }]
  }]
}

Add them to their team

Optionally group the joiner with their team to read the dynamic they are joining.

POSThttps://api.thomas.co/v1/teams

Request

POST /teams
{ "name": "Sales", "memberIndividualIds": ["00000000-0000-4000-8000-000000000022"] }

Response

{
  "id": "00000000-0000-4000-8000-000000000054",
  "tenantId": "00000000-0000-4000-8000-000000000001",
  "name": "Sales",
  "teamType": "explicit",
  "sourceManagerIndividualId": null,
  "visibility": "visible",
  "firstSharedAt": null,
  "needsOwnerAssignment": false,
  "deletedAt": null,
  "inactiveAt": null,
  "createdAt": "2026-07-02T10:00:00.000Z",
  "updatedAt": "2026-07-02T10:00:00.000Z",
  "members": [{
    "id": "00000000-0000-4000-8000-000000000091",
    "teamId": "00000000-0000-4000-8000-000000000054",
    "individualId": "00000000-0000-4000-8000-000000000022",
    "companyIndividualId": "00000000-0000-4000-8000-000000000032",
    "tenantId": "00000000-0000-4000-8000-000000000001",
    "role": "member",
    "status": "active",
    "createdAt": "2026-07-02T10:00:00.000Z",
    "updatedAt": "2026-07-02T10:00:00.000Z"
  }]
}

6
Recommend learning in your LMSyour system
Map the profile to onboarding paths in your LMS: an Action-led joiner gets short video plus a hands-on exercise, not long written courses.
7
Book intros in your calendaryour system
Schedule week-one 1:1s in your calendar and surface "how to work with your new teammate" in chat.
8
React to completion (webhook)
Coming soon
Result webhooks (an assessment.completed event) are coming soon. For now, poll the assignment status.

Build this with your agent

Open with Claude Open with ChatGPT

Open these calls in the API reference

When someone changes role or team, refresh how they fit.

1
Resolve the person
Resolve the person to one Thomas identity by email — the persistent identity key — or by your org-scoped HRIS id.
How identity works
POSThttps://api.thomas.co/v1/individuals/resolve
Request
```
POST /individuals/resolve
{ "email": "alex@acme.com" }
```
Response
```
{
  "individualId": "00000000-0000-4000-8000-000000000021",
  "created": false
}
```
2
Enrol as a colleague
Resolve returns a global identity with no tenant membership, so the profile read would reject it. Enrol the person as a colleague — createColleague resolves and enrols by email, as relationshipType employee — before you re-read their profile. The move preserves the same identity.
How identity works
POSThttps://api.thomas.co/v1/colleagues
Request
```
POST /colleagues
{ "email": "alex@acme.com" }
```

Re-read the profile

The profile persists across the move — same identity, by email or HRIS id. This reads only once the person has completed a Thomas behaviour assessment — something they do themselves; it cannot be triggered on their behalf through the API, and until then it returns 404 ("No scored assessment found for this individual").

How identity works

GEThttps://api.thomas.co/v1/assessments/profile/{individualId}

Request

GET /assessments/profile/00000000-0000-4000-8000-000000000021

Response

{
  "connectionPowerType": "energy",
  "traitSpectrums": [
    { "dimension": "D", "name": "Dominance", "leftPole": "Accommodating", "rightPole": "Direct", "position": 58, "band": "High" },
    { "dimension": "I", "name": "Influence", "leftPole": "Reflective", "rightPole": "Outgoing", "position": 64, "band": "High" },
    { "dimension": "S", "name": "Steadiness", "leftPole": "Spontaneous", "rightPole": "Methodical", "position": 41, "band": "Low" },
    { "dimension": "C", "name": "Compliance", "leftPole": "Pragmatist", "rightPole": "Perfectionist", "position": 38, "band": "Low" }
  ],
  "selfImageScores": { "d": 62, "i": 70, "s": 44, "c": 38 },
  "discAdvanced": { "workMaskScores": { "d": 58, "i": 66, "s": 49, "c": 41 }, "underPressureScores": { "d": 64, "i": 61, "s": 40, "c": 45 } },
  "scoredAt": "2026-10-02T09:30:00.000Z",
  "soloInsights": [{
    "topic": {
      "id": "00000000-0000-4000-8000-000000000052",
      "assessmentType": "behaviour",
      "slug": "collaboration-style",
      "label": "Collaboration style",
      "sortOrder": 1,
      "createdAt": "2026-09-01T00:00:00.000Z",
      "updatedAt": "2026-09-01T00:00:00.000Z"
    },
    "entries": [{
      "id": "00000000-0000-4000-8000-000000000062",
      "topicId": "00000000-0000-4000-8000-000000000052",
      "contentType": "markdown",
      "mappingKey": "energy-profile",
      "mappingGranularity": "profile",
      "title": "Keep the conversation lively",
      "body": "Use discussion and visible momentum to build alignment.",
      "sortOrder": 1,
      "createdAt": "2026-09-01T00:00:00.000Z",
      "updatedAt": "2026-09-01T00:00:00.000Z"
    }]
  }]
}

Read the new team's dynamic

Check the fit with the team they are joining.

GEThttps://api.thomas.co/v1/teams/{teamId}/dynamics

Request

GET /teams/00000000-0000-4000-8000-000000000041/dynamics

Response

{
  "teamId": "00000000-0000-4000-8000-000000000041",
  "members": [{
    "individualId": "00000000-0000-4000-8000-000000000021",
    "displayName": "Alex",
    "connectionPower": "action",
    "behaviourScales": {
      "accommodatingDirect": 72,
      "reflectiveOutgoing": 45,
      "spontaneousMethodical": 88,
      "pragmaticPerfectionist": 33
    },
    "assessmentStatus": "assessed"
  }],
  "powerDistribution": { "action": 1, "energy": 0, "stability": 0, "precision": 0 },
  "dominantPower": "action",
  "narrativeContent": null,
  "strengthsAndWatchouts": [{
    "type": "strength",
    "title": "Fast execution",
    "description": "The team moves quickly once aligned.",
    "narrativeContent": null
  }],
  "scaleDescriptions": null
}

Build this with your agent

Open with Claude Open with ChatGPT

Open these calls in the API reference

When someone leaves, capture how to work with them for the handover and gauge the team impact.

1
Resolve the person
Resolve the person to one Thomas identity by email — the persistent identity key — or by your org-scoped HRIS id.
How identity works
POSThttps://api.thomas.co/v1/individuals/resolve
Request
```
POST /individuals/resolve
{ "email": "alex@acme.com" }
```
Response
```
{
  "individualId": "00000000-0000-4000-8000-000000000021",
  "created": false
}
```
2
Enrol as a colleague
Resolve returns a global identity with no tenant membership, so the profile read would reject it. Enrol the person as a colleague — createColleague resolves and enrols by email, as relationshipType employee — before you read their profile for the handover.
How identity works
POSThttps://api.thomas.co/v1/colleagues
Request
```
POST /colleagues
{ "email": "alex@acme.com" }
```

Read their profile

Hand over "how to work with them" to whoever picks up their work. This reads only once the person has completed a Thomas behaviour assessment — something they do themselves; it cannot be triggered on their behalf through the API, and until then it returns 404 ("No scored assessment found for this individual").

How identity works

GEThttps://api.thomas.co/v1/assessments/profile/{individualId}

Request

GET /assessments/profile/00000000-0000-4000-8000-000000000021

Response

{
  "connectionPowerType": "energy",
  "traitSpectrums": [
    { "dimension": "D", "name": "Dominance", "leftPole": "Accommodating", "rightPole": "Direct", "position": 58, "band": "High" },
    { "dimension": "I", "name": "Influence", "leftPole": "Reflective", "rightPole": "Outgoing", "position": 64, "band": "High" },
    { "dimension": "S", "name": "Steadiness", "leftPole": "Spontaneous", "rightPole": "Methodical", "position": 41, "band": "Low" },
    { "dimension": "C", "name": "Compliance", "leftPole": "Pragmatist", "rightPole": "Perfectionist", "position": 38, "band": "Low" }
  ],
  "selfImageScores": { "d": 62, "i": 70, "s": 44, "c": 38 },
  "discAdvanced": { "workMaskScores": { "d": 58, "i": 66, "s": 49, "c": 41 }, "underPressureScores": { "d": 64, "i": 61, "s": 40, "c": 45 } },
  "scoredAt": "2026-10-02T09:30:00.000Z",
  "soloInsights": [{
    "topic": {
      "id": "00000000-0000-4000-8000-000000000052",
      "assessmentType": "behaviour",
      "slug": "collaboration-style",
      "label": "Collaboration style",
      "sortOrder": 1,
      "createdAt": "2026-09-01T00:00:00.000Z",
      "updatedAt": "2026-09-01T00:00:00.000Z"
    },
    "entries": [{
      "id": "00000000-0000-4000-8000-000000000062",
      "topicId": "00000000-0000-4000-8000-000000000052",
      "contentType": "markdown",
      "mappingKey": "energy-profile",
      "mappingGranularity": "profile",
      "title": "Keep the conversation lively",
      "body": "Use discussion and visible momentum to build alignment.",
      "sortOrder": 1,
      "createdAt": "2026-09-01T00:00:00.000Z",
      "updatedAt": "2026-09-01T00:00:00.000Z"
    }]
  }]
}

Read the team dynamic

See the gap their departure leaves in the team's balance.

GEThttps://api.thomas.co/v1/teams/{teamId}/dynamics

Request

GET /teams/00000000-0000-4000-8000-000000000041/dynamics

Response

{
  "teamId": "00000000-0000-4000-8000-000000000041",
  "members": [{
    "individualId": "00000000-0000-4000-8000-000000000021",
    "displayName": "Alex",
    "connectionPower": "action",
    "behaviourScales": {
      "accommodatingDirect": 72,
      "reflectiveOutgoing": 45,
      "spontaneousMethodical": 88,
      "pragmaticPerfectionist": 33
    },
    "assessmentStatus": "assessed"
  }],
  "powerDistribution": { "action": 1, "energy": 0, "stability": 0, "precision": 0 },
  "dominantPower": "action",
  "narrativeContent": null,
  "strengthsAndWatchouts": [{
    "type": "strength",
    "title": "Fast execution",
    "description": "The team moves quickly once aligned.",
    "narrativeContent": null
  }],
  "scaleDescriptions": null
}

Build this with your agent

Open with Claude Open with ChatGPT

Open these calls in the API reference

These are starting points, not the full set.