> ## Documentation Index
> Fetch the complete documentation index at: https://docs.b3os.org/llms.txt
> Use this file to discover all available pages before exploring further.

# Monitoring and Search

Observe workflows, runs, executions, activity streams, tags, templates, and searchable platform state.

B3OS gives teams visibility into workflow health through run history, execution details, activity streams, tags, templates, action details, and search-backed discovery.

**Operational debugging loop**

**Find run**
  **Inspect execution**
  **Check dependencies**
  **Repair draft**
  **Test and publish**
  **drill in**
  **isolate**
  **fix**
  **verify**

## What to Monitor

**Monitoring surface breadth**

**Signal count**

| Area | Signals |
| --- | --- |
| Workflow health | Active, paused, archived, repeated failures, validation errors |
| Runs | Status, trigger source, duration, CU usage, failure summaries |
| Executions | Node status, sanitized input/output, errors, retries, waits |
| Connectors | Expired credentials, permission errors, reconnect requirements |
| Wallets | Transaction submission, failures, nonce issues, gas or balance problems |
| Webhooks | Delivery status, retry count, replay history |
| Caddie | Recent sessions, repair attempts, and context quality |

## Activity Streams

Use streams for live UI and debugging experiences:

| Stream | Use |
| --- | --- |
| ` /v1/runs/:id/stream ` | Follow a single run |
| ` /v1/activity/stream ` | Follow workflow or organization activity |
| ` Execution streams ` | Inspect node-level progress and results |

## Searchable Objects

B3OS indexes platform objects so users and Caddie can find relevant resources.

| Object | Examples |
| --- | --- |
| Actions | Action IDs, titles, descriptions, tags, categories |
| Triggers | Trigger IDs, descriptions, and schema details |
| Workflows | Titles, descriptions, tags, status, visibility |
| Templates | Reusable workflow patterns |
| Organization knowledge | Approved org-specific context |
| Tags | Workflow and action discovery details |

## Debugging Flow

**Find the failed run**

Filter by workflow, trigger source, status, or time range.
  **Open the failing execution**

Inspect node input, output, sanitized error, and timing.
  **Check dependencies**

Review connector status, wallet funding, chain/provider availability, and action schema changes.
  **Repair the draft**

Fix expressions, missing fields, connector selection, or branch logic in a draft.
  **Test and publish**

Confirm the fix with a test run before publishing the new live version.

**Use tags for operations**

Tags make it easier to find workflows by protocol, customer, chain, risk level, owner, or operational area.