Copilot Agent: If a agent creates a new session with create_session the new session should be aware of the parent session

In UI we can clearly see that VS code has the parent and child link for each message send between the sessions but the agent itself is not aware of the fact that these messages are sent by a another session, for this I'm currently using this workaround by sending the , but agent some time make mistake in sending the correct uri.

This entire problem can also be solved in another way if the parent session auto receives a message on completion as it does for background subagents.

> How you are informing the child sessions that they are child session and they need to inform the parent session on completion. Because copilot doesn't expose this info to agent but I clearly see that vs code knows who is the parent.
>
> I'm using this workaround
>
> ```
> Call `create_session` with `relationship: "independent"` and `worktree: true` to create a worktree session for each implementer subagent, using this exact prompt and always keeping `/implement` in it:
>
> Rebase this branch onto , then /implement #.
> On completion, send exactly "# implemented on ." to using send_message.
>
> Here, `` is the latest commit after the most recent merge-agent commit, if one exists.
> ```
>
>
>
> And sending in prompt, agent sometime send wrong uri.
>
> I'm sharing my skill here, it is Matt Pocock implement-spec skill, that I have updated to use it with copilot.
>
> ```
> ---
> name: implement-spec
> description: "Implement a specification in code."
> ---
>
> You have been provided a spec. This spec should have tickets associated with it, describing how to implement the spec.
>
> The goal is a PR which implements the entire spec on a single branch.
>
> The tickets are not a list of steps. They are a **task graph** with blocking relationships between them. This means there is always a **frontier** of tickets which are ready to be grabbed.
>
> Communication to and from subagents should be sparse. Communicate primarily through **context pointers**: to the spec, tickets, research notes, and previous commits. Don't duplicate information already available via pointers.
>
> **Implementer subagents** should be run in the background where possible for **maximum concurrency**.
>
> After dispatching work, **end the turn** and resume only when a child sends its completion message.
>
> ## Steps
>
> 1. Read the spec and tickets. Read enough to understand the task graph.
>
> 2. (optional) Use an **exploration subagent** to conduct any exploration required by the tickets - relevant codebase files or external documentation. Ensure the exploration subagent can save files - it should save its markdown notes in a directory outside the repo, accessible by all future subagents. This lets **implementer subagents** focus on implementation rather than exploration.
>
> 3. Create a branch, and a draft PR. The PR should be marked as 'closing' the spec issue and tickets.
>
> 4. Use **implementer subagents** to implement each ticket. Each implementer subagent should work in its own worktree, on its own branch.
>
> 5. Once an **implementer subagent** completes, merge its work to the PR branch with a **merger subagent**.
>
> 6. If this changes the **frontier** of available tickets, kick off more **implementer subagents** to work on the new tickets. This allows for maximum concurrency.
>
> 7. Once all tickets are complete, run /code-review on the PR branch. Fix all issues raised by the code review in a single **implementer subagent**.
>
> 8. Mark the PR as ready for review.
>
> ## Implementer Subagent
>
> Call `create_session` with `relationship: "independent"` and `worktree: true` to create a worktree session for each implementer subagent, using this exact prompt and always keeping `/implement` in it:
>
> Rebase this branch onto , then /implement #.
> On completion, send exactly "# implemented on ." to using send_message.
>
> Here, `` is the latest commit after the most recent merge-agent commit, if one exists.
>
> ## Merger Subagent
>
> Create the first merge chat with `create_session` and `relationship: "currentSession"`, then reuse it with `send_message` for subsequent merges, always using this exact prompt:
>
> /merge into in this chat's current worktree as one commit named "Closes #: "
> On completion, send exactly "# merged." to this session's default chat using send_message.
>
> ## PR
>
> Do not create a draft PR in step 3. Create a ready-for-review PR in step 8 and mark it as closing the spec issue. Do not list tickets in the PR description.
> ```

_Originally posted by @MominRaza in [#335069](https://github.com/microsoft/vscode/issues/335069#issuecomment-5637216356)_