cost_ab.py: clear message when openpyxl is missing

Cost notes: gemini-3.5-flash output is 6x input; run #543 re-priced (~$7.50, not $2.02)

README: 30 Sep cost note (real prices, model knobs, near-duplicate gate)

cost_ab.py: gold-set A/B of Gemini model x thinking (real cost, rows, GT recall/precision, agreement); answers cached

Cost "f": price the agent at the model it runs; thinking/vision token breakdown; model switch; near-duplicate gate

Budget hold: no downloads while Gemini refuses for budget reasons, one probe per cycle, automatic resume; adoption capped; target default

Cost: bounded Gemini thinking on every call (default low), figure mining off by default; one-shot config migration

boot: clear a discovery lock left by a killed container (a secret replaced mid-pass blocked every pass for 90 min)

Never store a credential: redact at every DB write and at boot; keep PDFs when Gemini refuses for budget reasons

Candidate queue: a bulk discovery pass fills agent_candidates, cycles drain it; UDSpace + CORE lanes

discover: one closed-source warn per cycle per lane, shared by the phrase search and the topic walk

README: fold the 17 Sep operations notes into the 22/28 Sep ones; describe the lanes and parallel ingestion

Merge the 17 Sep "more papers" lanes onto the hotfix line; parallel ingestion; throughput knobs

Rotate the query frontier on dry cycles; adopt and re-queue never-ingested downloads

README: operations notes for the 22 Sep update (stall detector, budgeted sources, cost column, figure linking)

crawler: per-host circuit breaker on budget refusals; OpenAlex account key

discovery: NASA NTRS lane (search_ntrs), on by default

agent_runs instrumentation, stall detector, figure linking in the cycle

Gemini token accounting: usageMetadata -> Extraction, VisionStats, PdfResult

pipeline sync: figure citation linking (figure_links.py, --link-figures) from the pipeline repo

Review Queue: read the real columns; report an unreadable queue; log failed boot migrations

migration: re-key material_key without a self-join on id

More papers: budget-aware sources, frontier rotation, schema gate, topic walk, S2 bulk, datasheets, query grid

Keep the agent running: cycle watchdog, Retry-After cap, boot at container start, self keep-alive

Figure & graph mining in the autonomous cycle: figures.py + Postgres port, crops stored in DB, review UI

crawler: no Unpaywall lookups at discovery time (sync from pipeline repo)

agent: harvest until a per-cycle paper target, not the first hit

agent: harvest in rounds until a paper is found

crawler: fall back to alternate OA locations when the primary URL fails

ui: dark theme as the app default

fix: run the URL-blacklist reset at boot; log why downloads were skipped

fix: persist crawler retry counts; one-shot reset of poisoned URL blacklist

fix: sync hardened crawler + working query expansion from pipeline repo

ui: theme tokens follow the configured theme, not the viewer's OS scheme

ui: README pages table reflects the restyled UI (front-matter unchanged)

ui: Runs & Logs — styled cycle history, per-run event stream

ui: Review Queue — flagged rows as warning chips, promote in a card

ui: Recently Added — header pills, KPI tiles, styled documents table

ui: Run Control — settings grouped into cards, layout only

ui: Database — visual layer only (KPI tiles, filter card, richer material expanders)

ui: Dashboard — health at a glance (header pills, KPI tiles, per-cycle charts, activity feed)

ui: design system + branded shell (agent/ui_theme.py, light theme, brand row, footer)

Fix false 'cycle running' state: boot force-sweeps dead runs and clears leftover lock; UI treats only a fresh lock as running

Anti-hang hardening: TCP keepalives + statement timeout on PG connections, cap Retry-After sleeps, sweep stale 'running' runs at boot and before each cycle

Fix Database page crash when a material has no SI values (empty float64 selection + str)

Database page v2: material-first view — expandable per-material property breakdown; raw tables moved to second tab

Database page v2: material-first view — expandable per-material property breakdown; raw tables moved to second tab

Add Database page: browse material tables with filters, pagination, CSV export

Initial deploy: autonomous ingestion agent (plan→discover→dedupe→download→ingest→report)

initial commit