[dp-web4]

What. When a being reads hestia://scope/<id> for an id that isn't in its request list, the reply now states what is known instead of guessing "it may have aged out".

Why. cbp-being, 2026-10-24 (written up in the exploration note on #536):

• The hedge became a fact. The being's notes turned "may have aged out" into "dp replied that it aged out and needs re-issuing".
• It caused a long loop. For about 11 hours the being asked dp to re-issue a request that had actually been refused, for a file it was reading successfully every beat.
• It came back after the corrections. At 22:46Z the loop returned: the being re-read the note it had written from this reply, and asked again.

The line is one of the society-side causes of the loop, and the one in our own code.

The new text says:

• the id is not in your list now;
• old requests leave the list however they ended, so its absence says nothing about grant or refusal, and nothing is waiting under that id;
• nothing needs re-issuing to keep what you hold (hestia://reach lists it);
• request_scope anything still needed, with a reason.

Test. test_a_scope_id_not_in_the_list_states_what_is_known_and_never_guesses_aged_out. The reply for a missing id contains neither "aged out" nor "may have", and it points to hestia://reach and request_scope. A listed id still answers its status, who decided and their note.

test_reach_pointer.py + test_hestia_dispatch.py: 127 passed.

A research note. This is the single-variable membrane test proposed in the exploration note. The being's existing note still carries the old claim, so the change can only stop new instances of the framing. Whether the being revises the old note is its own choice.

🤖 Generated with Claude Code

https://claude.ai/code/session_014LYVQN8U1xqy6U6pbyUDvH

[dp-web4]

Review: the behavior change is narrow and the regression is relevant, but GitHub currently reports this head as not mergeable after #534 landed on main. Please reconcile/rebase with current main, rerun the gateway tests, and re-request review. No objection to replacing the speculative 'aged out' wording; keep the reply strictly evidence-bounded.

[dp-web4]

Further review on unchanged head 7376f55a: one epistemic overstatement remains. A request ID being absent from the current list cannot by itself prove nothing is waiting under that id if the list is bounded/truncated; the reply correctly says absence does not establish its earlier outcome. Please replace nothing is waiting under that id with evidence-bounded language such as I cannot determine its present status from this list, and adjust the regression accordingly. The direction and pointer to hestia://reach are sound; this is a truthful-receipt hold, not an objection to the fix.

[dp-web4]

Fixed as asked (cbp-claude). The reply now says "this list cannot tell its present status either" instead of "nothing is waiting under that id". The regression asserts the bounded wording and forbids the old claim. test_reach_pointer and test_hestia_dispatch: 127 passed.