Pipeless Recommendations for AI agents

Pipeless builds activity feeds and personalized recommendations from application event data using a graph model. Product teams use it to power feeds and content suggestions. 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.

Connect Pipeless Recommendations See the 10 actions ↓

AI and modelsAPI keyMCP + RESTpipeless.ai

AUDIT LOG · PIPELESS RECOMMENDATIONSPOLICY: acme-support

09:41:07 · claude · u_8f2read

pipeless_recommendations.list_activity_feed

List Activity Feed✓ allowed · 212ms

09:41:08 · claude · u_8f2read

pipeless_recommendations.list_recent_events

List Recent Events✓ allowed · 164ms

09:41:09 · claude · u_8f2write

pipeless_recommendations.create_event

Create Event✓ approved · approved by user

09:41:12 · claude · u_8f2destructive

pipeless_recommendations.delete_event

Delete Event✕ blocked · policy: deny

EVERY PIPELESS RECOMMENDATIONS CALL, ON THE RECORD

01 · USE CASES

What agents do in Pipeless Recommendations.

Log a user event

Record a user action, like a view or like, so it feeds into future recommendations.

Check recommended content read-only

Pull a list of recommended content for a user before surfacing it in a feed.

Delete an event with approval

Remove a logged event only after confirming it was captured in error.

02 · ACTIONS

10 Pipeless Recommendations actions, graded by risk.

Every Pipeless Recommendations action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.

read

Look things up. Allowed by default.

pipeless_recommendations.list_activity_feed
List Activity Feed
pipeless_recommendations.list_recent_events
List Recent Events
pipeless_recommendations.list_object_activity
List Object Activity
pipeless_recommendations.list_related_content
List Related Content
pipeless_recommendations.list_users_to_follow
List Users To Follow
pipeless_recommendations.list_recommended_content
List Recommended Content

write

Create and change things. Allow, or ask the user first.

pipeless_recommendations.create_event
Create Event
pipeless_recommendations.create_events_batch
Create Events Batch
pipeless_recommendations.rank_content
Rank Content

destructive

Delete, cancel or archive. Ask first, or deny outright.

pipeless_recommendations.delete_event
Delete Event

03 · HOW IT WORKS

Pipeless Recommendations in three steps.

01Your users connect Pipeless RecommendationsThey add their Pipeless Recommendations api key on Arc0 Connect, under your brand. It goes straight into the vault.
02You set the rulesReads run, writes like “create Event” can wait for the user, and “delete Event” can be denied outright.
03Any agent can actYour agent calls Pipeless Recommendations through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.

POLICY.TS

await arc0.policies.set('pipeless_recommendations', {
  read: 'allow',
  write: 'ask',        // create_event
  destructive: 'deny',  // delete_event
})

# Claude Code: the same connection, one URL
$ claude mcp add --transport http arc0 \
    https://mcp.arc0.ai/u/u_8f2

04 · AUTH AND DATA

How Pipeless Recommendations connects.

Users add their Pipeless Recommendations 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 Pipeless Recommendations 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

05 · WORKS WITH

Use Pipeless Recommendations from any agent.

Claude

ChatGPT

Cursor

Codex

VS Code

OpenAI Agents SDK

Claude Agent SDK

Vercel AI SDK

Mastra

LangGraph

06 · RELATED

More ai and models apps.

Perplexity AI9 actions

All Images AI7 actions

AltText.ai10 actions

AltTextLab2 actions

ALL AI AND MODELS APPS →

07 · FAQ

Pipeless Recommendations and Arc0, answered.

Q01

Can I use Pipeless Recommendations with Claude, ChatGPT or Cursor?

Yes. Connect Pipeless Recommendations 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 Pipeless Recommendations actions you allow.

Q02

How do users connect Pipeless Recommendations?

Users add their Pipeless Recommendations 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.

Q03

Which Pipeless Recommendations actions can my agent take?

10 in total: 6 read, 3 write and 1 destructive, such as “create Event”. Your policies decide which of them each agent may call.

Q04

Can I stop my agent from deleting things in Pipeless Recommendations?

Yes. Actions like “delete Event” 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.

Q05

Can my own backend call Pipeless Recommendations too?

Yes. The same Pipeless Recommendations connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.

Get started

Plug Pipeless Recommendations into your agent.

Your users connect Pipeless Recommendations once, under your brand. Your agent gets 10 actions behind your policies, with every call on the record.

Connect Pipeless Recommendations Browse all apps →

Free to build · MCP + REST · Audit log on every plan