Intent: the maintainer, 2026-11-03: "let's get the docs locked in", for the clients waiting on a clear specification. The reading battery's first run (ba9105b3cc), gemma4:31b on three checks twice, got 44 of 48 right, and each miss was the text's: in section 2.1 it read "with no info string or json" as refusing a fence opened with json, which src/ocx/messages.cr accepts and OCX itself writes; and in 2.12, an exit of 137 as surely signal 9, though the code writes 128 plus N for a signal as a shell does, so the two are written alike. A reading of the rest found more a client would trip on: 2.3 said a run with no sandbox runs unconfined while refusal 5 refuses it on a host without ROOT/settings/unconfined; 2.3 bounded runs from R by what the clone's owner allows, a file decision 132 removed; 2.8 wrote $release$ and $releases$ with one dollar each; and 2.11's filter 1 was refusal 1's row copied, which no code applies. Filter 5 also held back posts from any clone's user, which src/ocx/acting.cr's FILTERS never did, and the maintainer, 2026-11-03, of that constraint: "the constraint never made sense".

Expression: docs/protocol-v2.md: 2.1, an opening of three backticks alone or then json, spaces around json, and any other opening holding no message; 2.3, a run from R with the isolation refusal 1 asks, and an absent sandbox refused unless the host allows it (refusal 5); 2.8, $$release$$ and $$releases$$; 2.11, the filters as the code holds them: known, after, no message, not its own (the process's user alone), where, heard; 2.12, exited for a process ended by itself or a signal with no stop or limit, and an exit above 128 saying neither. 399 lines. battery/readings: section names may be the sections' anchors, as models cite them; clients.json adds a fourth check, on docs/protocol-v2-tools.md: messages written by other means, receipts, post's refusal, tolerant answers, a misplaced stop, ok and stopped, dry_run, and create from a checkout starting no thread.

Demonstrates: crystal run battery/main.cr -- --read all, gemma4:31b and qwen2.5-coder:7b, two trials each, on this text: gemma4:31b right on every question of every check, 64 of 64 (envelope, acting, runs, clients); qwen2.5-coder:7b 27 of 64 right, 23 partial, 14 wrong, its reasons often quoting the rule its numbered choice then contradicts, as the proposal's section 5 found of it. spec/battery/reading_spec.cr, spec/guards/documents_spec.cr, spec/guards/delineation_spec.cr, spec/guards/size_spec.cr, spec/guards/protocol_names_spec.cr and spec/guards/legibility_spec.cr: 18 examples, 0 failures.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01TUH8KLNdgQ5Qp4fkwUDDWT check-in: 1da76212f9 user: x86-1-session tags: protocol-v2-daemon

Intent: x86-1, sent by the maintainer after the switch to protocol version 2 at 10:25 PM Eastern on November 2: the server's create page refused every link links printed, "the run asks no user to act, and no actor acts on this host: give its login" (src/ocx/commands/create.cr), so x86-1 added login and clone to each link by hand; and the handoffs.json on the client held ten-digit run and handoff hashes, no asked and no stopped, so links printed no link at all and x86-1 filled it in by hand from each thread. Nobody should repeat either by hand.

Expression: ocx-rootfs/migrate-sessions.cr: list reads each watcher's acting clone's login, its default user as the daemon reads it with no user given (src/ocx/entry.cr), and its server-code, and Session and Handoff carry them as login and acting_clone (a record's clone is Crystal's copy). ask resolves the post it writes and the stop to full hashes, refusing an empty answer, an ambiguous prefix or no hash at all. links adds login and clone to each link, and for each session it gives no link names what it lacks, or which hash is not full, before its note. A run that ended lost after its stop was posted still counts: links reads the stop post, not the run's end. ocx-rootfs/README.md and the script's header say so, and no longer say list runs inside the client's namespaces, which the maintainer found unneeded. No fields override is added: patterns/session.md already binds no .claude.json and sets CLAUDE_CONFIG_DIR, since e7ef2dc409.

Demonstrates: ocx-rootfs/migrate-sessions.cr built; links, given a handoffs.json shaped like the client's at 9:39 PM, printed "ollama d9151401d0: no link, no stop, no login, no clone; not a full hash: run, handoff"; given full hashes with login x86-1 and clone fa361107a9, it printed the create link with login=x86-1 and clone=fa361107a96cf737cbba73269a20490cf7b227ff. On this run's clone, full resolved f3a9f7f433 to its 64 digits, gave nil for an empty answer and refused f; the login and server-code readers gave the clone's default user and server-code. patterns/session.md and patterns/run-claude.md, each a wiki page an admin wrote on a scratch repository, passed the live release's create --dry-run with login x86-1 and that clone. crystal run ci.cr: 646 examples, coverage 100.0%, its one failure spec/guards/checkin_spec.cr on e7ef2dc409's comment, which named spec paths short; that comment is amended and spec/guards/checkin_spec.cr passes, 8 examples, 0 failures.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01TUH8KLNdgQ5Qp4fkwUDDWT check-in: 276732f317 user: x86-1-session tags: protocol-v2-daemon