MCPBundler Docs
MCPBundler Docs
Back to marketplaceGetting StartedAuthenticationTutorial - Bundle to Connected Agent
API Reference
Agent A2A Tools
Agent Channels
Agent Configuration (Staging)
Agent Configuration (Versions)
Agent Tools
agents
Agents
Attachments
Auth
Billing
bundler
Bundler
Bundler Tool
Bundles
Call Campaigns
Catalog
Conversations
Credentials
Deployments
Discover
Fleet
Genesys
Insights
Acknowledge Insight IncidentArchive Insight TopicCreate Insight Alert RuleCreate Insight TopicDelete Insight Alert RuleGet Insight IncidentGet Insights SettingsInsight Sentiment MatrixInsight Processing StatusInsight KPIsInsight Time SeriesInsight Tool FailuresInsight Topic StatisticsInsight Config Version ComparisonList Insight IncidentsList Insight Alert RulesList Insight TopicsList Voice Engine PopulationsMerge Insight TopicUpdate Insights SettingsRe-analyse ConversationsResolve Insight IncidentInsight Topic Sample SummariesUpdate Insight Alert RuleUpdate Insight Topic
internal
Knowledge Base Website Sources
Knowledge Bases
MCP Listings
Media
OAuth Clients
OAuth2
Observability
Organisation roles
Organisations
Providers
Reviews
Roadmap
SCIM
Simulation Tests
Suppression
Triggers
TTS
Users
Utils
Visitors
Voices
Webhooks
Workflows
API ReferenceInsights

Re-analyse Conversations

POST
/v1/insights/reanalyse

Queue a fresh analysis of the organisation's closed conversations started within [since, until] (at most INSIGHTS_REANALYSE_MAX_DAYS), optionally for one agent. Pre-lifecycle conversations are closed into the lifecycle first. Emits no public event, and whether each job can run (credentials, credit) is decided when it is claimed.

One call per organisation per cooldown. A refused range, agent or candidate count does not consume the cooldown; once accepted the slot is spent whatever happens to individual conversations. Each conversation commits on its own, so at most one conversation lock is held at a time; one that fails is rolled back, logged and counted as failed. A retry after the cooldown is idempotent apart from creating new generations.

The work runs on the request thread, so a call accepts at most INSIGHTS_REANALYSE_MAX_CONVERSATIONS candidates (a size that finishes inside a proxy timeout); a larger backfill is split into narrower ranges or per-agent calls, one per cooldown.

Authorization

OAuth2AuthorizationCodeBearer
AuthorizationBearer <token>

In: header

Query Parameters

org_id*Org Id
Formatuuid
verify_locally?Verify Locally
Defaulttrue

Request Body

application/json

TypeScript Definitions

Use the request body type in TypeScript.

A re-analysis request over conversations started within [since, until]; a timestamp without a zone is read as UTC.

Response Body

application/json

application/json

curl -X POST "https://example.com/v1/insights/reanalyse?org_id=497f6eca-6276-4993-bfeb-53cbbbba6f08" \  -H "Content-Type: application/json" \  -d '{    "since": "2019-08-24T14:15:22Z",    "until": "2019-08-24T14:15:22Z"  }'
{  "queued": 0,  "legacy_closed": 0,  "skipped": 0,  "failed": 0}
{  "detail": [    {      "loc": [        "string"      ],      "msg": "string",      "type": "string"    }  ]}

Update Insights Settings

Previous Page

Resolve Insight Incident

Next Page

agent_id?|null
since*Since
Formatdate-time
until*Until
Formatdate-time