A brief that asks for a change closes Fixed by a default-branch commit, not Recorded on an open PR; work waiting on someone else is superseded by her own follow-up. Replaces the three 'Recorded with your answer' sentences.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
ChatDelivery gains tuple index 11, `image: Option<ChatImage>` ([path, mime,
size, sha256], a reference into her per-room image store, never bytes); the
pending kind is gamecult.persona.chat_pending.v2, and v1 answers and records
still decode with no image. chat_image::load is the one place a reference
becomes bytes: exact `<hex sha256 of room id>/<sha256>[.ext]` path with no
traversal or symlink, regular file, 5 MiB declared and actual, declared size
and sha256, png/jpeg/gif/webp magic bytes matching the declared mime. The
persona line carries the image as a base64 image block after its text and an
`attachment role=organ` line saying attached or not shown and why. The
stream-json echo of a line with blocks still counts as our echo. The store is
MIND_PERSONA_IMAGE_STORE; unset, every image is not shown.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
A DM's room is its room id under the reply cut (the steward cut's hand-built DM delivery carried the alias form); a request after a result opens a turn under the task cut, so the resume cut's bound test closes that turn with a result.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Cut: State.tasks, Event::TaskStarted, the tool_use_id keying of task lines and the notification replay, and TaskNotified's tasks.remove gate. running is a copy of the CLI's last background_tasks_changed, replaced whole; a notification owes a turn when the task was backgrounded and the session did not stop it with TaskStop (or KillShell, KillBash: the stream shows the alias as the model named it); Polled and Inbox carry other_first, so a served conversation's newer brief no longer outruns an older brief of another. Fixtures bg-fgto, bg-kill and bg-monitor pin the three falsified cases; the probe's fake API gains a generic tool step and a timed scenario.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
A foreground task moved to the background at its timeout, a task stopped with TaskStop, and a persistent Monitor; scrubbed to /probe/work, the key is the fake one.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
decide owns idleness (no turn open, nothing outstanding, no backgrounded task running, no notification owed its turn) and the switch of
conversation; read_usage reads the task lines for liveness only; the CLI's replay of a task notification is no echo of ours. SelfInbox
serves one Conversation and attributes a result by the span, not by the order of the stream. Deleted: briefless_open, briefless_answers,
turn_conversation, turn_allows, may_tell, one_conversation_told, note_task, tasks, notified, the Persona pointer line and its plumbing.
ScheduleWakeup and CronCreate are denied to the Self session. Tests replay the real-order captures with briefs spliced in.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The harness runs the CLI against the fake Messages API with a fake key and a throwaway config dir: bg and agent plan steps, a
Monitor step, --disallowedTools through DENY. Captured: an Agent call with run_in_background gives task_started
(is_backgrounded true, the call's tool_use_id) and a task_notification with the same id; a background Bash ending while no turn is open
makes the CLI begin its own turn (init, a request, no echo); ScheduleWakeup and CronCreate are absent from the init tools when denied;
a Monitor is a backgrounded task with the same two lines, so it needs no denial. Host paths scrubbed to /probe/work.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
A notification with no tool_use_id registers nothing (the open-turn gate still holds it by its request). Tests: the notified conversation's brief joins its turn, a notification of another conversation stays pending through a brief turn, two conversations' notifications open a turn no brief may join, an idle non-request line opens no turn, a resolved claim is read by its own key, a claimed cut spec stays on the start line.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
A request with no line of ours echoed is a turn that took in no brief, notification or not; until its result only the conversation it continues, or the notification's task's, is written. Nothing told yet means no other conversation exists.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The stream's task_started and task_notification events carry the tool_use_id of the call that started the task. The actuator records, per task, the briefs the starting turn was answering; a briefless result after a notification answers those (any notification it cannot place answers none). While a notification's turn is open, only the conversation it is attributed to may be written. A restart reads a resolved claim's conversation by key and opens the gate for it alone. The flush result is the note to the successor, in the journal: it is no reply. The start message states the rules.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
A reply to no brief is told in no session and stays in force (ruling briefless-result-routing). A restart drops a claim that is not of the start set conversation.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The start set is one conversation's; poll and tell gate briefs by the conversation of those unanswered;
a result that took in no brief continues the turn before; the start message says so.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
A flush written alone shows the people and open items the session's lines showed; which of
those are still open is read when a reply is decoded. A settle on a shown item that an
earlier turn closed is dropped with a trace note and the rest of the reply is admitted.
The flush no longer reads the mind at write time, so flush_text returns a String again.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
PersonaInbox.seen accumulates every person a write() showed; flush_text
becomes fallible and, when no line awaits its echo, sets Written for the
flush line from seen plus the open items and people read now.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The one conflict was the public-surface snapshot: the branch's new
launch::StopScope line beside main's LaunchOutcome::Tripped { run } line;
both kept.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Soul s1 scratch-symlink-deputy-write: she could plant a link at the
instructions file in the unit's RUNTIME_DIRECTORY and the organ's next write
followed it onto any path the unit's user can write, the hidden trace included.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The base is LF; the wip commits rewrote whole files as CRLF, hiding the real diff. Content unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
s5 findings of cut-self-session-live:
- blocked-recipe-misses-run-held-spec: eureka_state::hold is the one decider
(a question holds the item it is raised in, or every claim of the run it is
raised in; an answer of retire keeps holding the claims still in force).
queue::on_hold and `huginn blocked` both read it.
- hold-read-not-one-snapshot: on_hold returns a Held snapshot that the request
carries; claimable, queue and the gate read it, none reads it again.
- dead-self-holder-stalls-successor: launch::slot is the one reader of whether
a holder lives; claimable asks it, a dead holder is closed by the launch and
the beat's trace says so.
- retire-answer-reopens: launch::retire_answered resolves the claims of an
answered-retire question Withdrawn; the beat runs it before reading briefs.
- held-note-threshold-unpinned: a test pins the note for exactly one held brief.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Cut claimable's Mind-wide stop on any follow-up question and queue's inline
blocked loop. queue::on_hold is the one reader of what an in-force question
holds (a question raised in a run holds that run's claims, anything else holds
itself); queue, inbox::derive and the open_and_launch gate read it.
Breaker::Trip names the newest ended counted run; the breaker question is
raised in that run; the boundary is the newest of a question raised in the item
and a run claiming it that a question is raised in, so an answer or withdrawal
gives every claim of the run a fresh count. The trip and the held-brief wait
are written to the heartbeat trace.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>