docs: record the supervisor-as-second-writer mistake from RD-24

A task notification fires every time an agent stops with no live children,
including a pause mid-task. RD-24's agent paused and notified; the supervisor
read that as abandonment, took over the tree, ran the gate and edited the
ticket Status while the agent was still working. The agent resumed, correctly
detected a second writer, and refused to commit.

This is the mirror image of the RD-17 race the section above records: there a
fresh supervisor met a running agent, here a running agent met its own
supervisor.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

docs/project/readable-codebase/README.md

@@ -294,3 +294,18 @@ git log --oneline -1        # did an agent already commit the ticket?
 
 If the tree is dirty and this session did not dirty it, find the agent before you write a
 ticket around the evidence it leaves.
+
+**The mirror-image mistake, made during RD-24: the supervisor became the second writer.** A
+task notification fires every time an agent stops with **no live children — including a pause
+mid-task**. RD-24's agent paused, said it was waiting on a check, and notified. The supervisor
+read that as an abandoned task, took over the working tree, ran the gate and edited the ticket
+`Status:` — while the agent was still running. The agent then resumed, correctly detected a
+second process writing its tree, and refused to commit.
+
+Nothing was lost, because the agent stopped instead of committing. Two rules follow:
+
+- **A notification is not proof the work is finished.** A finished agent hands back a report
+  with acceptance numbers. "I am waiting for X" is a pause.
+- **Before taking over an agent's tree, confirm the agent is gone**: `ps aux | grep "npm run"`,
+  and check whether its own result has arrived. Taking over is a real option — RD-24 was
+  finished that way — but it must be a decision, not an assumption.