API DocsAnymize Features

Anymize Features

Create, manage and fetch projects, knowledge bases and meetings via the API — anymize's in-app features.

Beyond chat and anonymization, the API also drives anymize's in-app objects: projects (bundle chats, knowledge bases, a project instruction and a project memory), knowledge bases and Klartext meetings. Every endpoint works with your API key.

Concepts

Projects — bundle chats, knowledge bases, a project instruction and a project memory.
Knowledge Bases — knowledge bases you fill and attach to projects (RAG search).
Meetings — Klartext transcripts with speakers, timestamps and AI summaries.

Credits & billing

Read endpoints are free. The AI project-instruction endpoint (enhance-instruction) is billed on API-key calls at €2 per 1M input tokens and €5 per 1M output tokens; the credits spent are returned as credits_used.

Projects

Create a project

Creates a project and returns its ID.

POST/api/v1/projects

Parameter	Type	Default	Description
`project_name`required	`string`	-	Name of the project.
`project_instruction`	`string`	-	Optional project instruction (system instruction for every chat in the project).

1curl -X POST https://app.anymize.ai/api/v1/projects \
2  -H #86efac">"Authorization: Bearer YOUR_API_KEY" \
3  -H #86efac">"Content-Type: application/json" \
4  -d #86efac">'{ "project_name": "Onboarding", "project_instruction": "Antworte immer auf Deutsch." }'

201Created

{
  "success": true,
  "project": {
    "id": "proj_abc123",
    "project_name": "Onboarding",
    "project_instruction": "Antworte immer auf Deutsch.",
    "conversation_count": 0,
    "document_count": 0,
    "created_at": "2026-06-22T10: 30: 00Z"
  }
}

Fetch project data

Loads a project. With ?include= you get all chats (ID + name), linked knowledge bases and the project memory in one call.

GET/api/v1/projects/:id?include=conversations,knowledge_bases,memory

Parameter	Type	Default	Description
`include`	`string`	-	Comma-separated: conversations, knowledge_bases, memory.

200OK

{
  "project": {
    "id": "proj_abc123",
    "project_name": "Onboarding",
    "project_instruction": "Antworte immer auf Deutsch.",
    "conversations": [
      {
        "id": "conv_1",
        "chat_name": "Kickoff",
        "message_count": 12
      }
    ],
    "knowledge_bases": [
      {
        "id": "kb_1",
        "name": "Handbuch",
        "rag_status": "completed"
      }
    ],
    "memory": "# Projektgedächtnis\n- Kunde bevorzugt E-Mail."
  }
}

Manage: name, instruction & databases

Set name and project instruction via PATCH (no AI). Link knowledge bases via the knowledge-bases endpoint.

PATCH/api/v1/projects/:id

POST/api/v1/projects/:id/knowledge-bases

1# Name + Projektanweisung setzen (ohne KI)
2curl -X PATCH https://app.anymize.ai/api/v1/projects/proj_abc123 \
3  -H #86efac">"Authorization: Bearer YOUR_API_KEY" \
4  -H #86efac">"Content-Type: application/json" \
5  -d #86efac">'{ "project_name": "Onboarding 2026", "project_instruction": "Kurz und freundlich." }'
6
7# Wissensdatenbank verknüpfen
8curl -X POST https://app.anymize.ai/api/v1/projects/proj_abc123/knowledge-bases \
9  -H #86efac">"Authorization: Bearer YOUR_API_KEY" \
10  -H #86efac">"Content-Type: application/json" \
11  -d #86efac">'{ "knowledgeBaseId": "kb_1" }'

Project memory

Read, set or delete the project memory (markdown).

PUT/api/v1/memory/project/:projectId

1curl -X PUT https://app.anymize.ai/api/v1/memory/project/proj_abc123 \
2  -H #86efac">"Authorization: Bearer YOUR_API_KEY" \
3  -H #86efac">"Content-Type: application/json" \
4  -d #86efac">'{ "content": "# Projektgedächtnis\n- Kunde bevorzugt E-Mail." }'

Project documents

Lists all documents/jobs that belong to a project.

GET/api/v1/projects/:id/documents

Knowledge Bases

Create & fill a knowledge base

Creates a knowledge base and returns its ID. Upload files afterwards.

POST/api/v1/knowledge-bases

1# 1) Wissensdatenbank erstellen
2curl -X POST https://app.anymize.ai/api/v1/knowledge-bases \
3  -H #86efac">"Authorization: Bearer YOUR_API_KEY" \
4  -H #86efac">"Content-Type: application/json" \
5  -d #86efac">'{ "name": "Produktwissen", "description": "Alle Handbücher" }'
6
7# 2) Dateien hochladen (multipart)
8curl -X POST https://app.anymize.ai/api/v1/knowledge-bases/kb_1/upload \
9  -H #86efac">"Authorization: Bearer YOUR_API_KEY" \
10  -F #86efac">"file=@handbuch.pdf"

201Created

{
  "success": true,
  "knowledgeBase": {
    "id": "kb_1",
    "name": "Produktwissen",
    "rag_status": "pending",
    "doc_count": 0
  }
}

Document status

Shows per-document processing and embedding status (embedded, pending_embedding, processing, failed).

GET/api/v1/knowledge-bases/:id/documents

200OK

{
  "documents": [
    {
      "id": "job_1",
      "filename": "handbuch.pdf",
      "status": "completed",
      "embedding_status": "embedded",
      "chunk_count": 42
    },
    {
      "id": "job_2",
      "filename": "faq.docx",
      "status": "completed",
      "embedding_status": "pending_embedding",
      "chunk_count": 0
    }
  ],
  "total": 2
}

Link to a project

Connects a knowledge base to a project.

POST/api/v1/projects/:id/knowledge-bases

AI project instruction (billed)

Turns a short description into a structured project instruction. Billed on API-key calls (€2/1M input, €5/1M output tokens).

POST/api/v1/projects/enhance-instruction

1curl -X POST https://app.anymize.ai/api/v1/projects/enhance-instruction \
2  -H #86efac">"Authorization: Bearer YOUR_API_KEY" \
3  -H #86efac">"Content-Type: application/json" \
4  -d #86efac">'{ "transcript": "Ein Assistent fuer unser Support-Team, freundlich und kurz." }'

200OK

{
  "success": true,
  "instruction": "Du bist der Support-Assistent …",
  "credits_used": 137
}

Meetings

List meetings

Lists your Klartext meetings and recordings — lightweight rows without the transcript (id, title, date, duration, word count).

GET/api/v1/meetings

Parameter	Type	Default	Description
`limit`	`number`	`100`	Maximum count (default 100, max 200).

200OK

{
  "data": [
    {
      "id": "job_1",
      "title": "Team-Sync 15.07.",
      "created_at": "2026-07-15T09: 00: 00Z",
      "words": 1840,
      "is_anonymized": false,
      "duration_ms": 1860000,
      "has_audio": true,
      "kind": "meeting"
    }
  ]
}

Fetch a meeting

Loads a meeting in full: job basics, transcript (text + segments with speaker and timestamps) and every finished summary. Audio bytes are not included.

GET/api/v1/meetings/:id

1curl https://app.anymize.ai/api/v1/meetings/job_1 \
2  -H #86efac">"Authorization: Bearer YOUR_API_KEY"

200OK

{
  "job": {
    "id": "job_1",
    "title": "Team-Sync 15.07.",
    "kind": "meeting",
    "language": "de",
    "words": 1840
  },
  "transcript": {
    "text": "Guten Morgen zusammen …",
    "segments": [
      {
        "speaker": "S1",
        "start_ms": 0,
        "end_ms": 5100,
        "text": "Guten Morgen zusammen …"
      }
    ]
  },
  "summaries": [
    {
      "template_key": "general",
      "title": "Team-Sync 15.07.",
      "content": "## Zusammenfassung …",
      "locale": "de",
      "status": "completed"
    }
  ]
}

Jobs & Documents

List jobs

Lists your jobs — uploaded documents, text anonymizations, URL imports and transcriptions — with filters for project, status and search term.

GET/api/v1/jobs

Parameter	Type	Default	Description
`projectId`	`string`	-	Only jobs in this project.
`status`	`string`	-	Only jobs with this status (pending, processing, completed, failed).
`search`	`string`	-	Full-text filter on title/filename.
`limit`	`number`	-	Maximum count per page.
`offset`	`number`	-	Offset for pagination.

1curl #86efac">"https://app.anymize.ai/api/v1/jobs?status=completed&limit=25" \
2  -H #86efac">"Authorization: Bearer YOUR_API_KEY"

Fetch a job

Loads a single job with status, metadata and project assignment — e.g. to pick up the result of an anonymization.

GET/api/v1/jobs/:id

200OK

{
  "data": {
    "id": "job_1",
    "title": "Vertrag_2026.pdf",
    "filename": "Vertrag_2026.pdf",
    "status": "completed",
    "is_anonymized": true,
    "language": "de",
    "words": 5230,
    "project_id": "proj_1",
    "created_at": "2026-07-10T08: 00: 00Z"
  }
}

Create a text job

Creates a job from text — optionally with anonymization (withAnonymization) and linked straight to a project and/or a knowledge base. Anonymized jobs start as 'pending'; poll completion via GET /api/v1/jobs/:id.

POST/api/v1/jobs/upload-text

1curl -X POST https://app.anymize.ai/api/v1/jobs/upload-text \
2  -H #86efac">"Authorization: Bearer YOUR_API_KEY" \
3  -H #86efac">"Content-Type: application/json" \
4  -d '{
5    #86efac">"text": "Hallo, mein Name ist Max Mustermann ...",
6    #86efac">"title": "Support-Ticket 1234",
7    #86efac">"withAnonymization": true,
8    #86efac">"projectId": "proj_1",
9    #86efac">"knowledgeBaseId": "kb_1"
10  }'

201Created

{
  "data": {
    "id": "job_1",
    "status": "pending"
  },
  "job_id": "job_1",
  "status": "pending"
}

Create a URL job

Imports the content of a website or a YouTube video (transcript) as a job — optionally straight into a knowledge base. Note: URL imports are not anonymized.

POST/api/v1/jobs/upload-url

1curl -X POST https://app.anymize.ai/api/v1/jobs/upload-url \
2  -H #86efac">"Authorization: Bearer YOUR_API_KEY" \
3  -H #86efac">"Content-Type: application/json" \
4  -d #86efac">'{ "url": "https://example.com/artikel", "knowledgeBaseId": "kb_1" }'

File uploads

Anonymize files (PDF, DOCX, …) via POST /api/anonymize; plain text extraction runs via POST /api/ocr — see the Anonymization tab. Poll progress via GET /api/status/:id or GET /api/v1/jobs/:id.