Reflect

Read the installed poteto-mode reference ../poteto-mode/references/t3-execution.md before selecting models or delegating. It owns named-role resolution, catalog option keys, accounts and automatic authorized missing-usage fallback.

Mine the current conversation for durable learnings, then route them into skill edits.

Invoke when the user says "reflect" or "/reflect". Skip when the conversation is trivial, off-topic, or already covered by an existing skill the parent followed correctly. One-offs are not learnings.

1. Locate the active transcript

Before fanning out, find this conversation's thread in the active project with t3_thread_list and read it with t3_thread_read. Do not search across unrelated projects. That crosses workspace boundaries and reads private chats from unrelated projects.

Use the current thread ID from t3_thread_list and match its opening user prompt through t3_thread_read. Page the activity and recover truncated item text until the full conversation is available.

If the active thread is unavailable or unreadable, pass a tight digest of the session instead.

2. Spawn three reviewers in parallel

One message, three delegate_task calls with mode: "async" and self-contained briefs. Reviewers need MCP access to look up the tickets, chat threads, and observability traces the thread references. Keep that access while constraining edits in each brief.