Jira source

Use kind = "jira" to collect Jira issues with JQL.

[[sources]]
id = "jira-ready"

[sources.source]
kind = "jira"
site = "https://example.atlassian.net"
jql = "project = AB AND statusCategory = Todo"
limit = 50

Credentials come from environment variables by default. Defaults are JIRA_EMAIL and JIRA_API_TOKEN.

[sources.source]
kind = "jira"
site = "https://example.atlassian.net"
email_env = "JIRA_EMAIL"
token_env = "JIRA_API_TOKEN"
jql = "project = AB ORDER BY updated DESC"

Or configure a credential helper. AgentBoard writes Git-style request lines to stdin (protocol, host) and reads either username/password or email/token lines from stdout.

[sources.source]
kind = "jira"
site = "https://example.atlassian.net"
jql = "project = AB ORDER BY updated DESC"

[sources.source.credentials]
helper = "agentboard-jira-credentials"

Field mapping

Defaults map id from key, title from fields.summary, status from fields.status.name, and url to {site}/browse/{key}. Override mappings when you need custom Jira fields. Use status_map to normalize Jira status names.

AgentBoard automatically requests Jira fields referenced by mapping paths that start with fields.. For example, fields.customfield_10010.value requests the top-level Jira field customfield_10010.

[sources.source.field_map]
id = "key"
title = "fields.summary"
status = "fields.status.name"
url = "fields.customfield_10010"

Use status_map when Jira's status names should normalize to AgentBoard statuses.

status_map = { "To Do" = "ready", "In Progress" = "doing" }

Use fields only for extra Jira API fields you want preserved in the raw payload but do not use in field_map.

fields = ["assignee"]