vv_is_ai_host() was `=== 'host1'`, which made a physical fact — that is where the
GPU is — look like a rule. AI_OWNER_HOST declares it, so the card can move to a
rebuilt host3 or a friend's spare.
The gate was also doing two jobs. Assistant docks and findings strips now ask
whether a model is reachable, so a node without a GPU gets them by borrowing;
the AI tab asks whether this is the owner, because that page carries the bug
reports, the index and the model configuration — the surface where the
vocabulary assumes you built the mesh.
Resolution is local, then owner, then anyone else declaring a model, pinned once
it answers. Pinned rather than re-derived per call: a mesh that re-decides every
request eventually decides differently mid-conversation, and a chat whose second
turn lands on another machine has no history there. Cleared only on a transport
failure, and only when there is somewhere else to go — a single-node mesh whose
model is down should say so, not report AI as unconfigured.
The triage could say Narcos is filed under anime and offer nothing to do about
it. Findings that recorded a destination now carry a Move action, which the AI
tab renders as a button on its own — actions are read from the server, so no UI
knew this existed. Armed like Fix and Dismiss, because it is the only action in
that card that relocates files rather than settings and the undo is another move.
Offered only where the triage said misfiled, never for uncertain: three
narrowings before a file is touched. Re-reads the series from Sonarr first, so a
finding gone stale cannot undo a move made by hand in the meantime.
Search is the one capability that fits the profile holding none: every other capability either
reads this installation or changes it, and chat's whole contract is that it can do neither. The
assistant deliberately does not get it — its contract is that answers come from this install's
own documents, and a web result there is an answer that looks sourced and is not. A chat question
about this machine hands off to the assistant before the search would run, so it never reaches
the internet even with the box ticked.
Off by default, and not because it is dangerous: searching sends the operator's question outside
the house, which is theirs to decide. Asked for per turn as well as enabled in conf.
Provider-agnostic, as asked — searxng, brave, tavily. Only whichever is configured here can be
verified; all three read every field by name, so a shape that changes underneath yields no
results rather than wrong ones.
The explain fixtures asserted chat holds no capabilities at all, which is exactly the guarantee
worth keeping. caps=only: now states the set rather than its emptiness, so the check still fails
the day something else is granted there.
Seven findings had been sitting in needs_operator with no way to reach the operator: the only
way to read one was the sweep tool over SSH. The card states the gate as well as the list,
because an empty list means nothing found when repair is on and nothing is looking when it is
off. Buttons come from the actions the endpoint returned for that row, so a tab left open
overnight cannot offer a choice the store has moved past.
A stored transcript is replayed into a later prompt when reopened, so a key typed while
changing a setting would be handed back to the model on every subsequent turn.
The Tools card adopted the AI scripts on the host check alone, and api/ai.php only
tested AI_ENABLED in front of ask, so a disabled subsystem still had rows to run and
an endpoint that answered.
Explicit buttons rather than an automatic router. Misclassifying a Varaverk
question as chat produces a confident invention about the user's system, which
is exactly what retrieval exists to prevent — with buttons there is no hidden
heuristic to be wrong and the strict profile is the default you land on.
Only Varaverk Assistant retrieves; the other two would be carrying passages
that cannot help write a folder-copy script. Memory goes to all three, since
that is what lets chat know the setup without claiming authority over it.
History depth is per profile and set server-side: retrieval costs ~2500 of
16384, so the profiles that skip it can hold a real conversation. Code
Sketcher is told to flag flags it is unsure of, after it invented
rsync --no-overwrite.
A small operator-written file handed to the model at the start of every
conversation — who you are, how this install is set up, what has already been
decided. Injected ahead of the retrieved passages and marked operator-authored
so it outranks anything they contradict, and never cited as a source.
Deliberately not indexed and deliberately under DATA_DIR: it changes
constantly, vector similarity is the wrong way to retrieve things you were
told to remember, and gitignoring it keeps personal notes out of a pushed
repository. The character cap is a context budget — this text costs its share
of 16k on every single turn.
A POST the browser demonstrably sent — correct token, correct body — left no
trace: no CSRF termination, no fatal, and no action log. Rejected-before-PHP
and died-inside-the-include are indistinguishable from outside, so the first
statement now records that execution arrived.
There is no nginx access log on this host and the CSRF prepend exits with an
empty body, so a request that never arrived and one that arrived and failed
were indistinguishable — the page just sat at "starting…". The job-file write
was also suppressed with @, which would produce exactly that hang: a token
returned for a job that can never report.
Token and poll rather than SSE, so the api layer keeps one response
convention and reuses the pattern manual_sync already proved. History is
capped at three turns because the model is only fully offloaded at 16384
context and unbounded history would cross that silently. The tab exists
only while AI_ENABLED is true, rejected server-side and not merely hidden.
Gmer4Lfe