SafetyCulture for AI agents
SafetyCulture, formerly iAuditor, is a workplace operations platform where frontline teams run digital inspections, audits, and safety checklists, and track assets and corrective actions. Connect it once through Arc0, and your agent, or Claude, ChatGPT and Cursor, can use it through one MCP endpoint, limited to what each user approved.
What agents do in SafetyCulture.
Create a corrective action from an inspection
Opens a new action item with an owner and due date after a digital inspection flags a hazard.
Look up an asset's inspection history
Pulls an asset's field records and audit trail before deciding whether it needs servicing.
Archive an asset only after confirmation
Retires an asset record from the register only once a supervisor confirms it is decommissioned.
255 SafetyCulture actions, graded by risk.
Every SafetyCulture action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.
read
129Look things up. Allowed by default.
- safetyculture.get_assetGet Asset
- safetyculture.list_sitesList Sites
- safetyculture.list_typesList Asset Types
- safetyculture.list_usersList Users
- safetyculture.get_actionsGet Actions
- safetyculture.get_webhookGet Webhook
- safetyculture.list_assetsList Assets
- safetyculture.list_fieldsList Fields
- safetyculture.list_groupsList Groups
- safetyculture.list_issuesList Issues
- safetyculture.get_timelineGet Timeline
- safetyculture.list_actionsList Actions
- safetyculture.list_groups2List Groups (Feed)
- safetyculture.list_sensorsList Sensors
- safetyculture.get_documentsGet Documents
- safetyculture.get_asset_typeGet Asset Type
write
97Create and change things. Allow, or ask the user first.
- safetyculture.update_typeUpdate Type
- safetyculture.create_assetCreate Asset
- safetyculture.create_groupCreate Group
- safetyculture.update_assetUpdate Asset
- safetyculture.update_fieldUpdate Field
- safetyculture.create_actionCreate Action
- safetyculture.create_documentsCreate Documents
- safetyculture.update_documentsUpdate Documents
- safetyculture.create_asset_typeCreate Asset Type
- safetyculture.create_user_fieldCreate User Field
- safetyculture.update_user_fieldUpdate User Field
- safetyculture.create_asset_fieldCreate Asset Field
- safetyculture.create_assets_bulkBulk Create Assets
- safetyculture.update_assets_bulkBulk Update Assets
- safetyculture.update_action_assetUpdate Action Asset
- safetyculture.add_file_owners_bulkBulk Add File Owners
destructive
29Delete, cancel or archive. Ask first, or deny outright.
- safetyculture.delete_assetDelete Asset
- safetyculture.archive_assetArchive Asset
- safetyculture.delete_groupsBulk Delete Groups
- safetyculture.delete_actionsDelete Actions
- safetyculture.delete_webhookDelete Webhook
- safetyculture.delete_incidentsDelete Incidents
- safetyculture.delete_asset_typeDelete Asset Type
- safetyculture.delete_asset_fieldDelete Asset Field
- safetyculture.remove_file_ownersBulk Remove File Owners
- safetyculture.delete_users_fieldsDelete User Field
- safetyculture.archive_document_itemArchive Document Item
- safetyculture.delete_tasks_task_typeDelete Action Type
- safetyculture.delete_directory_foldersDelete Directory Folders
- safetyculture.delete_tasks_custom_fieldDelete Tasks Custom Field
- safetyculture.delete_incidents_categoriesDelete Incident Category
- safetyculture.delete_incidents_osha_casesDelete Incidents OSHA Case
SafetyCulture in three steps.
- 01Your users connect SafetyCultureThey add their SafetyCulture api key on Arc0 Connect, under your brand. It goes straight into the vault.
- 02You set the rulesReads run, writes like “create Asset” can wait for the user, and “delete Asset” can be denied outright.
- 03Any agent can actYour agent calls SafetyCulture through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
await arc0.policies.set('safetyculture', { read: 'allow', write: 'ask', // create_asset destructive: 'deny', // delete_asset }) # Claude Code: the same connection, one URL $ claude mcp add --transport http arc0 \ https://mcp.arc0.ai/u/u_8f2
How SafetyCulture connects.
Users add their SafetyCulture api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.
The same SafetyCulture connection serves your agent over MCP and your own backend over REST and the proxy, so a user connects once. How Arc0 handles credentials →
- AUTH
- API key
- CREDENTIALS
- Per-tenant encrypted vault
- MODEL SEES
- Results only, never credentials
- AUDIT LOG
- Every call, on every plan
Use SafetyCulture from any agent.
ClaudeChatGPTCursor
Codex
VS CodeOpenAI Agents SDKClaude Agent SDK
Vercel AI SDK
Mastra
LangGraph
SafetyCulture and Arc0, answered.
Can I use SafetyCulture with Claude, ChatGPT or Cursor?
Yes. Connect SafetyCulture to Arc0 once, then add your Arc0 MCP URL to Claude, ChatGPT, Cursor, Claude Code or any other remote-MCP client. Each assistant only gets the SafetyCulture actions you allow.
How do users connect SafetyCulture?
Users add their SafetyCulture api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.
Which SafetyCulture actions can my agent take?
255 in total: 129 read, 97 write and 29 destructive, such as “create Asset”. Your policies decide which of them each agent may call.
Can I stop my agent from deleting things in SafetyCulture?
Yes. Actions like “delete Asset” are graded destructive. Set destructive actions to deny, or to ask so the user approves each one, and blocked calls still show up on the audit log.
Can my own backend call SafetyCulture too?
Yes. The same SafetyCulture connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.
Plug SafetyCulture into your agent.
Your users connect SafetyCulture once, under your brand. Your agent gets 255 actions behind your policies, with every call on the record.
Free to build · MCP + REST · Audit log on every plan