Dify · Agentic Access

Dify Agentic Access

x-agentic-access generated

Dify exposes 13 API operations that an AI agent could call, of which 10 are state-changing ‘acting’ operations. This is a recommended x-agentic-access execution contract — the scope, audience, consequence tier, short-lived token constraints, and escalation each action should carry before it is handed to an autonomous agent.

By consequence: 3 read, 5 write, 2 physical, and 3 safety-critical.

3 operations are classed safety-critical and should require human-in-the-loop approval at runtime.

Contracts are classified heuristically from the provider’s OpenAPI and refresh on every APIs.io network build; audience is bound per deployment. The model follows Curity’s Access Intelligence (apidays Munich 2026). Browse every provider’s agent contracts at agentic-access.apis.io.

Artificial IntelligenceBackend-As-A-ServiceLLMOps

Operations: 13 Acting: 10 Human-in-the-loop: 3 Method: generated

By consequence

read 3 write 5 physical 2 safety-critical 3

Highest-consequence actions

The physical and safety-critical operations an agent could invoke — the ones that most warrant scoped tokens, tight TTLs, and escalation. Full per-operation contracts are in the source below.

Method	Path	Consequence	Human-in-loop
POST	/chat-messages/{task_id}/stop	safety-critical	required
POST	/completion-messages/{task_id}/stop	safety-critical	required
POST	/workflows/{task_id}/stop	safety-critical	required
POST	/chat-messages	physical	conditional
POST	/completion-messages	physical	conditional

Source

Agentic Access

generated: '2026-07-15'
method: generated
source: openapi/dify-openapi.yml
description: Recommended x-agentic-access execution contracts, classified heuristically from
  the OpenAPI. A governance starting point for exposing this API to AI agents — review and bind
  audience per deployment. See research/curity/agentic-governance/.
summary:
  operations: 13
  by_action_class:
    acting: 10
    connected: 3
  by_consequence:
    physical: 2
    safety-critical: 3
    write: 5
    read: 3
  human_in_the_loop_required: 3
operations:
- path: /chat-messages
  method: post
  operationId: sendChatMessage
  x-agentic-access:
    action-class: acting
    consequence: physical
    subject: required
    audience: null
    token:
      max-ttl: 300
      exchange: true
      purpose-required: true
    escalation:
      human-in-the-loop: conditional
      triggers:
      - abnormal
      - high-value
    audit: required
- path: /chat-messages/{task_id}/stop
  method: post
  operationId: stopChatMessage
  x-agentic-access:
    action-class: acting
    consequence: safety-critical
    subject: required
    audience: null
    token:
      max-ttl: 120
      exchange: true
      purpose-required: true
      proof-of-possession: true
    escalation:
      human-in-the-loop: required
    audit: required
- path: /completion-messages
  method: post
  operationId: sendCompletion
  x-agentic-access:
    action-class: acting
    consequence: physical
    subject: required
    audience: null
    token:
      max-ttl: 300
      exchange: true
      purpose-required: true
    escalation:
      human-in-the-loop: conditional
      triggers:
      - abnormal
      - high-value
    audit: required
- path: /completion-messages/{task_id}/stop
  method: post
  operationId: stopCompletion
  x-agentic-access:
    action-class: acting
    consequence: safety-critical
    subject: required
    audience: null
    token:
      max-ttl: 120
      exchange: true
      purpose-required: true
      proof-of-possession: true
    escalation:
      human-in-the-loop: required
    audit: required
- path: /workflows/run
  method: post
  operationId: runWorkflow
  x-agentic-access:
    action-class: acting
    consequence: write
    subject: required
    audience: null
    token:
      max-ttl: 900
    escalation:
      human-in-the-loop: conditional
      triggers:
      - abnormal
      - high-value
    audit: required
- path: /workflows/{task_id}/stop
  method: post
  operationId: stopWorkflow
  x-agentic-access:
    action-class: acting
    consequence: safety-critical
    subject: required
    audience: null
    token:
      max-ttl: 120
      exchange: true
      purpose-required: true
      proof-of-possession: true
    escalation:
      human-in-the-loop: required
    audit: required
- path: /workflows/run/{workflow_run_id}
  method: get
  operationId: getWorkflowRun
  x-agentic-access:
    action-class: connected
    consequence: read
    subject: optional
    token:
      max-ttl: 3600
    audit: none
- path: /conversations
  method: get
  operationId: listConversations
  x-agentic-access:
    action-class: connected
    consequence: read
    subject: optional
    token:
      max-ttl: 3600
    audit: none
- path: /conversations/{conversation_id}
  method: delete
  operationId: deleteConversation
  x-agentic-access:
    action-class: acting
    consequence: write
    subject: required
    audience: null
    token:
      max-ttl: 900
    escalation:
      human-in-the-loop: conditional
      triggers:
      - abnormal
      - high-value
    audit: required
- path: /conversations/{conversation_id}/name
  method: post
  operationId: renameConversation
  x-agentic-access:
    action-class: acting
    consequence: write
    subject: required
    audience: null
    token:
      max-ttl: 900
    escalation:
      human-in-the-loop: conditional
      triggers:
      - abnormal
      - high-value
    audit: required
- path: /datasets
  method: get
  operationId: listDatasets
  x-agentic-access:
    action-class: connected
    consequence: read
    subject: optional
    token:
      max-ttl: 3600
    audit: none
- path: /datasets/{dataset_id}/retrieve
  method: post
  operationId: datasetRetrieve
  x-agentic-access:
    action-class: acting
    consequence: write
    subject: required
    audience: null
    token:
      max-ttl: 900
    escalation:
      human-in-the-loop: conditional
      triggers:
      - abnormal
      - high-value
    audit: required
- path: /files/upload
  method: post
  operationId: uploadFile
  x-agentic-access:
    action-class: acting
    consequence: write
    subject: required
    audience: null
    token:
      max-ttl: 900
    escalation:
      human-in-the-loop: conditional
      triggers:
      - abnormal
      - high-value
    audit: required