memory: one search across facts, raw history and sketches ...

Compression removes messages from the model's window, not from the
conversation: every compression path now writes the turns it demotes into
`memory_chunks` — verbatim, append-only, with the turn range, timestamps and
breadcrumb keys (the summarizer's `KEYS:` line plus a deterministic extractor,
which also covers the hard-truncate fallback that never sees a summarizer).

`memory_search` is the single read path over the three stores: curated facts
first, then raw excerpts by relevance, then the current session's sketches.
Every hit carries provenance (source, session, turn range, timestamp), and a
raw excerpt that a newer curated fact overrides is labelled outdated rather
than hidden — "we changed our mind, the old value was X" is often the
question. The agent also gets up to two relevant excerpts injected per turn
once the context holds a summary, deduped across turns.

Fixes found by running the pilot against a real database:
- `search_facts` returns ISO-string timestamps while chunks return datetimes;
  the freshness comparison raised TypeError and took the whole search down.
- an existing `memory_chunks` table created without pgvector never gained the
  `embedding` column (CREATE TABLE IF NOT EXISTS does not alter), and the
  INSERT failed on it; the DDL now adds the column, and a database without
  pgvector writes text-only rows found through FTS.

context: price images by area and detect backend-truncated prompts ...

Every window decision priced an inline image at a flat 500 tokens, while a
1024x1024 screenshot costs ~1026 on qwen3.6 and a 2048x2048 one ~4098. The
undercount kept sessions below the 90% trigger; Ollama then silently truncated
the prompt to about half the window (dropping the system prompt), and the
truncated prompt_eval_count was cached as the session baseline, pinning the
estimate at ~50% for good - the session never compressed while the model was
already told the context was full.

- image_tokens.py prices each image from its header (one token per 32x32 px,
  downscaled to <=2048^2, clamped to 256..4096); unreadable headers fall back
  to CONTEXT_IMAGE_TOKENS. Used by both estimators, the delta and
  check_context_size; the per-turn image budget fits images by their own cost.
- A prompt reported far below what was sent is a truncation: it is not cached
  as the baseline, and the post-turn worker compresses regardless of the
  threshold. The session keeps the honest token count.
- The pre-turn gate prices the incoming message and its images, which are not
  in session.context yet.
- The summarizer receives images under a window-scaled budget, and prompts
  carrying images stay out of the calibration fit (their bytes are not in
  chars and would skew it).
- Oversized user messages are capped for the LLM view (at least num_ctx/3),
  so a paste bigger than the window no longer dead-ends a session.
- The in-flight partition also trims by token budget: a gate that fires can
  now actually shrink a turn made of a few huge messages.

sessions: share a chat as a public read-only link ...

A session's owner can hand out one capability link that opens the display
history without an account: no input, no chat list, no artifacts panel.
The token lives in the URL hash, so it never reaches the server in a request
line or a Referer header.

The link is permanent by decision - no expiry, no revoke. Deleting the chat
removes the share row (FK cascade) and closes it, which is the only switch.

Backend: session_shares table (unique token, FK cascade), POST
/sessions/{id}/share issuing the session's one link (owner or admin; a
read-all permission may read another user's chat but not publish it), and
public GET /shared/{token}[/messages] serving display history only, with
session and owner ids left out of the payload.

Webclient: #share=<token> renders a standalone page with the owner's cards,
rating off, interactive UI components filtered out and scroll-up paging; a
share button in the chat header opens a dialog that creates the link.

Also fixes ChatHeader's rename, which called an undefined `api` and failed
silently.

llm: rotate the routing config under a running process ...

A key that lives in the central config is only worth centralizing if
publishing it does not cost a fleet restart. The backend bakes host and
Authorization header into its clients, so it now accepts a fresh
LLMConfig: clients are dropped and rebuilt on the next call, health
stays keyed by target name, and the probe follows the new hosts too.

The overlay's llm section is no longer always staged. It is re-parsed
against the local document and swapped live; only a backend without the
swap method or a document that fails to parse falls back to
restart_required.

overlay: ack the config again after a restart ...

A document that needed a restart was acked as 'staged, restart required'
and never updated afterwards: the machine restarted into the new config but
the hive kept showing the old ack. Startup now re-reports what it is running,
so the status view reflects reality instead of the last live refresh.

overlay: report llm and mcp changes as restart_required ...

A live refresh cannot swap LLM backends or the MCP pool — both are built
with the container — so a document that changes those sections was reported
as applied while the old wiring kept running. Compare the sections against
the previous document and stage them by name.

navi: overlay the hive config onto the instance's own ...

An instance that knows a hive with the swarm key starts from the hive
document: layers that apply to it are merged over its local config, fields
the hive does not mention stay local, and an empty answer means 'run exactly
your own config'. Settings, profiles, LLM routes and MCP servers are covered;
database URL, ports, auth switches and the channel's own settings can never be
overridden.

Startup fetches with a short timeout, falls back to the last validated overlay
on disk and refuses to crash-loop on a document that killed the previous start.
A live refresh applies secrets and timeouts in place (the frozen singleton is
written through one deliberate escape hatch), rebuilds profiles from their
local base and reports everything else as restart_required. The hive reaches
the instance by push, by the announce reply and by a retry sweep; every apply
is acked back, which is what the hive's status view reads.

profiles: teach the agent to scout profiles and tools before switching ...

Persona gains a reconnaissance order (list_profiles, then list_tools for the
real toolset, then tool_manual) and the switch-vs-delegate rule: switch when
the task itself is another profile's work, delegate a side piece with
spawn_agent. The profiles block is built in one place and shared with the
admin prompt debug view, and the stale 'next user message' claim is fixed.

list_tools applies the admin-only role gate to an explicitly named profile,
the same one list_profiles uses, so it cannot enumerate hidden profiles.

skills: re-authorize a replacement when the proposal is accepted ...

A proposal proved ownership when it was written, but the protocol can be
promoted to shared while it sits pending. Check again at the click: a shared
protocol is now admins only, a personal one belongs to its owner.

webclient: read the skills catalog as documents, split it into two tabs ...

Settings had one long panel mixing skills and protocols as raw markdown,
which was hard to read and wasted the width. Skills and Protocols are now
two tabs over one panel keyed by kind.

Selecting an entry shows the parsed document — triggers, preconditions,
steps, post-checks, escape hatches, body — and an Edit button switches to
the source, so the catalog is read the way the agent loads it. The
single-entry payload carries the parsed document; the list still does not,
so it stays cheap. SkillDocument is shared with the chat proposal card,
which now only draws its own frame and header.

skills: name the kind when a document already had one implied ...

A proposal is parsed as a protocol when it is stored, but accepting it
re-parsed the same text with no override, so a front matter that left
'kind' out made the catalog refuse a document the user had just agreed
to. The create path and the agent's own skill writes had the same gap.

prompt: give the methodologies catalog its rules in the system prompt

profiles: give every working profile the skills and protocol tools

protocol: run a protocol as a committed plan, with a log of how it ended ...

A protocol was only text until now. The new `protocol` tool starts one: its
steps are spliced into the session's todo plan after the agent's current
position (nothing already done is reordered), a row appears in `protocol_runs`
with a snapshot of the version, steps and post-checks, and the run stays
`running` until it is closed.

`finish` refuses while any step of the run is open, refuses outright when one
failed, and requires a report — it is the run's record for the user. `cancel`
and `fail` require a reason and mark the rest of the run's steps `skipped`, so
the plan does not keep work that was never done. One run per session is active
(a partial unique index enforces it).

The commitment also has to survive the context: every model turn while a run is
open carries a system reminder with the protocol, its steps and the way to close
it, and the final-turn intercept now fires for an open run even when the todo
shows no open steps, so a bare-text turn cannot leave one running.

`protocol` is wired into assistant, developer, discuss and server_admin next to
`skills`; sub-agents get neither, because a protocol belongs to the session's
plan. The run log has no UI yet — `protocol(op='status')` is the current view.

skills: agent-written skills and protocol proposals with chat buttons ...

A skill the agent writes now lands in the catalog on its own; a protocol
never does. `skills(op="propose")` stores the document in `skill_proposals`
and returns a card through the tool result's `ui_component` metadata — no
`render_component` call, so profiles without `navi_ui` get the card too.

The decision is the server's: the browser sends `skill_proposal` over the
WebSocket, `SkillStore.decide_proposal` writes the catalog row from the
*stored* document, names the clicking user as the author, and answers with
`skill_proposal_decided`. The agent learns the outcome as a hidden user
message, the same path a `navi_ui` form submission takes.

`create`/`update` refuse protocol documents and point at `propose` with
`replaces_id`, so the version history of a protocol always starts from a
user's yes. The chat store now renders `ui_component` from any tool whose
result carries one, not only from `mcp__navi_ui__render_component`.

skills: catalog of skills and protocols with a settings tab ...

One registry for reusable methodologies: markdown with YAML front matter,
stored in PostgreSQL alongside a full edit history. A row is either a
`skill` (advice, adapt as needed) or a `protocol` (a committed order of
actions), and a protocol is refused on save unless it declares its steps,
post-checks and the three escape hatches — on_deviation, on_blocked,
on_already_done — because an honest run needs a way out when reality
differs from the plan.

- `SkillStore`: lazy DDL, uniqueness on (owner_id, kind, name), version
  bump with an appended `skill_versions` row, `content_hash` as the only
  hive-sync hook. Personal rows carry an owner; shared rows have an empty
  one and are writable by an admin.
- REST `/skills` (list/get/create/update/delete/versions) and the agent
  tool `skills` (search/load) that renders a compact document — numbered
  steps and escape hatches for a protocol, the instructions for a skill.
- Settings → Скилы и протоколы: catalog with search and kind/area
  filters, a document editor that shows the server's validation error
  verbatim, create from a template, delete, edit history, and the
  personal/shared switch for an admin.

Manuals stay where they are: tool knowledge, separate from this catalog.

todo: give plans their own table and recover them after a crash ...

Todo already survived restarts inside the generic session_store KV row, but
the plan there was opaque JSON with no timestamp: after a server crash the
agent resumed a step marked in_progress with no way to tell whether it had
ever been done.

session_todos (user_id, session_id, tasks JSONB, revision, updated_at) now
holds the plan. Existing rows are migrated on first use and the KV source is
deleted in the same step, so clearing a plan is not undone by the next
process start.

Each step gains provenance fields (source, step_id, protocol_run_id) for the
protocol work, and insert_steps() splices a block after the step the agent
has already reached instead of making it re-issue the whole plan.

get_progress_message() now flags a plan whose in_progress step predates this
process once per (session, revision): re-check the real result before
marking it done.

The todo op "clear" now uses the same user scope as load/save; it used to
prefer ctx.user_id and could delete a different row than the one written.

export: token-served age snapshot of dialogue for an n-gram consumer ...

Navi had no way to hand its history to another machine. A database dump
was the obvious idea and the wrong one: prod's 597 MB is mostly things
that must never leave (Fernet-encrypted MCP keys, user rows, attachments),
and the dialogue an n-gram table wants is 14 MB of it.

So the export is a fixed, versioned SELECT — role, content, session, seq,
ts — over both message tables. Default corpus is user+assistant only:
tool output is 52 MB of scraped web text and command logs, and system
messages are near-duplicate boilerplate; either would dominate the counts.
A rolling window takes whole sessions, newest first, up to 150 MB, so the
snapshot stays bounded as history grows.

The snapshot is age-encrypted to a recipient whose private half never
reaches this server, so Navi cannot decrypt what it exports. The plaintext
gzip is deterministic, and its digest becomes content_id: an unchanged
corpus keeps the ciphertext and its ETag, so a consumer that polls every
few minutes downloads nothing until something actually changed.

Endpoints sit behind a bearer token (machine credential, not user auth):
/export/v1/manifest describes, /export/v1/messages streams and answers 304
on a matching If-None-Match. /health carries an informational freshness
object that never moves the aggregate status.

scripts/navi_export.py is the consumer side: keygen, then pull — manifest
first, download, checksum, decrypt, gzip-verify, atomic rename; exit 3
when nothing changed.

subagents: a stopped run closes the tool it had opened ...

A tool card is drawn from tool_started and only its tool_call closes it.
When the user stopped a run while a sub-agent was mid-tool, the sub-agent
was cancelled, the parent dropped its sink, and the matching tool_call was
never sent — the card spun until the page was reloaded.

The sub-agent now reports its in-flight tool from the CancelledError
handler, and both batch paths (sequential and parallel) drain the sink
after cancelling instead of discarding it, so that event reaches the
client ahead of the parent's synthetic "stopped by the user" result.

The webclient settles anything still pending at stream_end,
stream_stopped and error as a safety net for a lost frame or a dying
server, with a visible "interrupted" note rather than a silent close.

The TUI needs nothing: it renders tool_call as a result panel and carries
no pending state, so it gets the fix through the backend alone.

subagents: a worker persona, a live prompt fallback, and guards that scale ...

- assistant and designer_3d get their own subagent_system_prompt.txt; without
  one a sub-agent ran on the orchestrator's prompt, or since ce55325 on nothing
  but the scope anchor, and rarely finished the step it was given
- build the sub-agent system prompt in one place; the documented fallback to
  profile.system_prompt had become dead code
- the thinking-stall guard killed healthy cloud reasoners at 12k chars after
  13-35s; raise the char bound to 48k, keep the 60s wall, and make both settings
- a 300s run cap cut long ops work in half; raise it to 600s, configurable

webclient: a recall you can inspect, skip and cancel ...

Clicking the banner opens a dialog with the call's kind, when it was created,
a live countdown to the next fire, the repeat interval and the system message
the agent wrote for itself; skip and cancel live there too.

The cancel button in the banner and the sidebar never worked: ph-clock-slash
is not a Phosphor icon, so the glyph was empty and the button collapsed to
8x4px. It is ph-x-circle now, with a larger hit area.

/recall payloads carry created_at, and the countdown picks days or hours
instead of counting out "43230 min".

sessions: the info dialog shows a pending recall ...

GET /sessions/{id}/info now carries the session's pending recall, and
GET /sessions/{id}/recall answers in one shape ({"recall": {...} | null})
instead of a flat object when found and a wrapper when not. The lookup is a
single indexed row rather than the newest 50 recalls filtered in Python.

compression: keep the last four turns verbatim, not eight ...

Four turns still carry the current task word for word; everything older is
already summarised, so eight bought little and pushed the compression trigger
later than it needed to be.

The mid-turn window is unchanged: it counts messages, not turns, and
max(12, keep_recent*2) still floors at 12. navi_code keeps its own override of
12 turns.

Examples, docs and the two tests that assert the default follow.

compressor: count the context the model is sent, not the history it never sees ...

The session-info modal showed "817 288 / 131 072 (100%)" for a session whose
context held a 1.47 MB code_exec result. The estimate was honest about
session.context, but build() head/tail-truncates every oversized tool/assistant
message to context_message_token_budget before the call, so the model never
received that megabyte.

truncate_oversized() moves the per-build rule to module level (the method now
delegates, behaviour unchanged) so the compressor can price the same view it
will send: compress_and_save_session stores estimate_tokens(truncate_oversized(
new_context)) instead of estimate_tokens(new_context).

Full content stays in stored history — the user still sees it, reloads are
unaffected. Regression test keeps a 1 MB tool result in the context and asserts
the stored count equals the truncated-view estimate and is >10x smaller than
the raw one.

webclient: a dialog that answers which model is serving, and from where ...

The header only named a model when the route had fallen back, so "what is
running this session, on which host, with how much context" had no answer
in the UI. The context percentage doubles as the second way in: a session
bar is a fair place to ask what the bar measures.

GET /sessions/{id}/info reads in-process state only — the active route
entry, the health chain, the profile's effective window, run and client
counts — and the dialog polls it while open. /meta and the session head
now report the route's window instead of the global ollama_num_ctx, which
had been wrong for every cloud-backed profile.

llm: a key by any name, so a second cloud account needs no code ...

`_resolve_secret` knew exactly two variables: `OLLAMA_API_KEY` and
`EMBEDDING_OLLAMA_API_KEY`, because those were the only ones with a Settings
field to bridge to. Any other name — a second Ollama Cloud account on the
backup chain — was looked for in `os.environ` only, and pydantic-settings does
not export `.env` there, so the target silently received an empty key and
every call to it would 401.

The lookup now reads the `.env` file itself for the name it was given, in the
same CWD-relative way pydantic-settings resolves it, with `os.environ` first
and the two known Settings fields last. `python-dotenv` was already a
dependency; a malformed line in the file (production has one) does not stop
the variables below it from being found, and that is a test.

`llm.json` stays secret-free: each target keeps naming its variable, several
accounts are just several targets, and the docs spell out the chain.

llm: routes that name a host, a backup and a window, and a profile that reads them ...

`llm.json` replaces `OLLAMA_BACKENDS_FILE`. It has targets (host and key
environment variable) and routes, where a route is an ordered chain of
`model @ target` entries, each with its own `context_window` and options. The
`default` route now starts on `glm-5.3-flash:cloud` and backs it with
`qwen3.6:35b-128k` on the home host, both at 131072 tokens, so a failover no
longer changes the budget the compressor and `num_ctx` are accounted against.
Embeddings get their own `embedding` route instead of a pinned host.

The engine classifies a failure (network, quota, auth, server, missing model),
cools that entry and moves the call down the chain; a background probe watches
only degraded targets and lets the next turn return to the primary once it is
healthy — it never interrupts a running turn. A profile joins a route with
`llm_route`, and the loader now actually reads it: `llm_route` and
`context_window` were declared on `AgentProfile` and written into every
`config.json` while the loader skipped both, the same silent no-op that
`is_admin_only` used to be, so every profile ran the legacy model list at the
global window. The writer emits them too, or an admin edit would erase the
route from disk.

The turn's resolved entry travels to the UI: the first chunk is stamped with
model and target, `model_info` carries `fallback` when a backup answered, and
the chat header shows a chip only then. `serving_primary` compares the target
as well as the model name, because a route may run one model on two hosts.
`/health` gains an `llm` block with the active entry, target states and missing
models.

`deploy/ollama/` keeps the local model's definition honest: MTP off (slower at
every context on this host), `presence_penalty 0` (the base model's 1.5 breaks
agentic tool calls), and `num_gpu 40` — Ollama's automatic fit claims a 128k
f16 slot fits and then dies mid-prefill with CUDA OOM; 40 of 42 layers leaves
~400 MiB and decodes at ~77 tok/s, 38 layers collapses to 48, and 39 OOMs.
q8_0 KV measured 37 tok/s against 70 on f16 here, so the cache stays f16.

profiles: the Personal Secretary retires, and every reference goes with it ...

`secretary` had become a second general-purpose assistant: same models, same
tools, one temperature step apart from `assistant`, and admin-only for no
reason the user could see. The directory is gone and the eight remaining
profiles load clean.

A profile that no longer exists is not a harmless orphan — `Agent` resolves
`session.profile_id` on every run and raises `ProfileNotFound`, so any session
left behind would open and then fail on its first turn. Prod sessions were
moved to `assistant` first (115 rows, verified after), and the local clone
along with them (101).

The reaction fallback loses its role table: it existed because the admin
fallback was itself admin-only, which `assistant` never is. `spawn_agent`
falls back to `assistant` too, and the two prompts that recommended
`secretary` now name only `server_admin` and `developer`.

navi: a token count that matches the model, a type for every file, and a socket that admits it is dead ...

Compression decided everything in chars // 3, and on this workload that sits
~1.8x below reality — 101k real prompt tokens against a ~56k estimate — so the
trigger, the target, the truncate budgets and the pre-call size check all ran
in a unit half the size of the real one. TokenCalibration now fits
overhead + chars / chars_per_token per model from the prompt_eval_count that
every call already reports, and a profile declares its own context_window
(sent to Ollama as num_ctx) instead of every model being held to OLLAMA_NUM_CTX.

Superseded summaries were never demoted in the database: save() only updates
flags for rows in session.messages, a summary is is_display=False, so it lived
in session.context alone and every reload resurrected it — prod sessions
carried up to 15 live summaries. PgSessionStore.set_context_flag() writes the
flip for those rows directly.

Five extension tables disagreed about what a file is. navi/content_types.py is
the one detector — signature, then extension, then a text probe — it decides
the MIME the download route serves, and the session file list now carries the
type the viewers branch on.

The idle receive loop answers client pings and sends its own heartbeat. A
half-open socket keeps readyState at OPEN and never fires close, so traffic
coming back is the only proof the connection still works.

secrets: a placeholder may carry a fallback, and web search stops being a Settings field ...

An unset ${SEARXNG_URL} failed the whole navi-web connection, all three of its
tools and not only search, because resolve_secrets raises on any placeholder it
cannot fill. The rule is right for credentials, where a blank value quietly
becomes anonymous access, and wrong for a setting, where an empty SearXNG URL
costs one backend rather than the server. ${NAME:-fallback} (POSIX-ish, empty
fallback allowed) says which is which, and writing the fallback down makes it a
decision in the config instead of an accident of a variable nobody set.

Settings.brave_search_api_key and .searxng_url were read by nothing — the
navi-web server has its own config — so they go, with docs/config.md pointing at
mcp_servers.d/navi-web.json and deploy/env.template naming SEARXNG_URL where a
deploy actually sets it.

The literal-secret guard still holds: a placeholder name is a credential unless
it is in SETTINGS_VARIABLES, and a settings placeholder must sit in an env block.