Taggun for AI agents
Taggun reads receipts and invoices and turns the image into structured line-item data, used by finance and expense teams to automate the manual side of expense tracking. 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 Taggun.
Transcribe a receipt image
Extract merchant name, date, and line-item totals from a photographed or scanned receipt.
Process a receipt from a URL
Pull structured expense data from a receipt hosted at a URL instead of an uploaded file.
Confirm a merchant mapping before saving
Check a new merchant name against existing records before adding it, so future receipts categorize correctly.
14 Taggun actions, graded by risk.
Every Taggun action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.
read
3Look things up. Allowed by default.
- taggun.export_known_merchantsExport Known Merchants
- taggun.export_product_categoriesExport Product Categories
- taggun.export_known_product_codesExport Known Product Codes
write
11Create and change things. Allow, or ask the user first.
- taggun.add_merchant_nameAdd Merchant Name
- taggun.urlProcess Receipt via URL
- taggun.url_verboseURL Verbose
- taggun.url_validationURL Validation
- taggun.generate_merchants_csvGenerate Merchants CSV
- taggun.import_known_merchantsImport Known Merchants
- taggun.import_product_categoriesImport Product Categories
- taggun.import_known_product_codesImport Known Product Codes
- taggun.transcribe_receipt_file_simpleTranscribe Receipt File (Simple)
- taggun.transcribe_receipt_encoded_simpleTranscribe Receipt from Base64 Encoded Image
- taggun.transcribe_receipt_encoded_verboseTranscribe Receipt Encoded Verbose
destructive
0Delete, cancel or archive. Ask first, or deny outright.
- No destructive actions.
Taggun in three steps.
- 01Your users connect TaggunThey add their Taggun api key on Arc0 Connect, under your brand. It goes straight into the vault.
- 02You set the rulesReads run, and writes like “add Merchant Name” can wait for the user to approve.
- 03Any agent can actYour agent calls Taggun through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
await arc0.policies.set('taggun', { read: 'allow', write: 'ask', // add_merchant_name destructive: 'deny', }) # Claude Code: the same connection, one URL $ claude mcp add --transport http arc0 \ https://mcp.arc0.ai/u/u_8f2
How Taggun connects.
Users add their Taggun 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 Taggun 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 Taggun from any agent.
Taggun and Arc0, answered.
Can I use Taggun with Claude, ChatGPT or Cursor?
Yes. Connect Taggun 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 Taggun actions you allow.
How do users connect Taggun?
Users add their Taggun 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 Taggun actions can my agent take?
14 in total: 3 read, 11 write and 0 destructive, such as “add Merchant Name”. Your policies decide which of them each agent may call.
Can I make my agent read-only in Taggun?
Yes. Allow read actions and deny writes in the Taggun policy. Your agent can still look things up, and any write it attempts is blocked and logged.
Can my own backend call Taggun too?
Yes. The same Taggun connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.
Plug Taggun into your agent.
Your users connect Taggun once, under your brand. Your agent gets 14 actions behind your policies, with every call on the record.
Free to build · MCP + REST · Audit log on every plan