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

# Follow the fan-out: delegation gets observable

> Each issue an orchestrator delegates now runs in its own session and Slack thread, and a new tool reports the live status of every delegated agent.

*August 26, 2026*

Two upgrades make agent-to-agent delegation easier to follow:

* **One thread per delegated issue**: when an orchestrator fans work out (one ticket per failing model, for example), each delegated issue now runs in its own session with its own Slack thread, instead of stacking every update into one thread.
* **Ask how the fan-out is going**: the new `check_delegated_agents` tool reports the live status and latest output of every agent a session delegated to with `invoke_agent`, so an orchestrator can answer "is anything still running" mid-run without waiting on callbacks.

See [agent-to-agent delegation](/products/dino-ai/programmable-agents/agent-to-agent-delegation) and the [agent tools reference](/products/dino-ai/programmable-agents/tools-reference). Shipped with the August 26, 2026 platform release.


## Related topics

- [Agent-to-Agent Delegation](/products/dino-ai/programmable-agents/agent-to-agent-delegation.md)
- [Agent Sessions](/products/dino-ai/mcp-server/agent-sessions.md)
- [Jira backlog agent](/guides/programmable-agents/jira-backlog-agent.md)
- [Linear backlog agent](/guides/programmable-agents/linear-backlog-agent.md)
- [Query cost optimizer (BigQuery)](/guides/programmable-agents/dbt-tm-model-query-cost-optimizer-bigquery.md)