Briefings (AI briefing of what happened) (part 2)
How a briefing is actually assembled: the sources it draws on, the pass that turns them into prose, and the schedule it runs on.
On this page
What it is
The continuation of part 1, covering the engine behind a briefing rather than the reader-facing recap.
Where to find it
The briefing runs on its own schedule; the code that assembles it is under the daily-digest service named in the frontmatter of this page.
How it behaves
How it works
Gathering runs off the main thread (2026-07-13). Building a digest scans a whole day of sessions, messages, and channel activity in the database. Those scans used to run synchronously on Omniscio's main window thread — and when Windows had pushed the database out of RAM under heavy disk load, they could freeze the entire window for minutes at a time, once per 15-minute retry. The scans now run in a dedicated short-lived database helper process (amc-db-host-digest, spawned per digest and always terminated afterward), so a slow disk can stall the digest — never your window. If the helper can't start, gathering falls back to the old in-process path; a mid-scan failure just fails that cycle and retries at the next 15-minute tick. Escape hatch: the AMC_DISABLE_DIGEST_OFFTHREAD=1 environment variable forces the legacy path. (The same fix repaired a silent day-one bug: the "recent project history" context the AI uses for feature ideas had never actually loaded — digests generated after this change include it for the first time.) Behavior lock: digest-gather-offthread-contract.md.
The prompt stays bounded on busy days (2026-07-16). A single very active day can involve hundreds of sessions. The digest used to list every one of them in the prompt it sends to Haiku — and once the "recent project history" fix above started actually loading, that block grew with your project count too. On the busiest days the request grew large enough that the model couldn't finish inside the call's time budget, so it timed out, and the automatic catch-up retried the same oversized request every cycle — leaving those days permanently blank (this is what stranded 2026-07-12/13/14). Two caps fix it at the root: the prompt enumerates at most the 120 most-recent sessions (the stats row still counts the true total, and a "showing the 120 most recent of N" note tells the model it saw a subset) and at most the 10 most-active projects in the history block — mirroring the per-channel message caps every other source already had. The generation call also now uses the digest's own ~50-second budget instead of silently inheriting the provider's shorter 30-second default. Together these keep even a heavy day's request well inside the time budget, so catch-up can fill it. Caps are MAX_SESSIONS_IN_DIGEST / MAX_HISTORY_PROJECTS in digest-prompt-builder.ts, locked by daily-digest-prompt-bounds.test.ts.
The service is daily-digest-service.ts — runs a 15-minute createPeriodicTask loop, checks settings.dailyDigestGenerationHour against local time, and if a digest for yesterday is missing it pulls source data via the enabled integrations (each source has its own helper: session-summary, gmail-summary, calendar-summary, …), concatenates them, and sends the whole thing through llmProviderService.chat() with provider: 'openrouter' + responseFormat: 'json_object'. The router uses the embedded OpenRouter key from default-credentials.local.json for the cheap path (so the digest works without any user-configured Anthropic account), falling back to the user's Anthropic apikey only if the cheap provider 5xxs or returns empty content. Generation skips only when there is no usable credential of either kind (no embedded OpenRouter key AND no user apikey); that genuine can't-generate case — plus an OpenRouter-only generation failure with no apikey to fall back on — raises ONE de-duplicated inbox alert ("Couldn't generate your daily digest") so the otherwise-silent stop surfaces instead of digests just disappearing. Every generation failure is routed exactly once (routeDigestFailure, 2026-09-08) into one of three surfaces, and the same verdict sets the log level so the two can't disagree: a credential/billing failure (rejected key, exhausted monthly allowance, no key left to fall back to, dry credit balance) raises that actionable card; a transient failure (a timeout/abort, a 429/5xx, a bare connection error — a machine that is offline, where nothing reached the server — or AiExtrasUnavailableError — every row in the AI extras key order skipped because its key is paused after failures or the allowance cannot serve right now) touches nothing, logs at warn and self-heals on the next 15-minute tick once the breaker cools; only a genuinely unexplained failure feeds the "your digest keeps failing" give-up streak. Before this, only an HTTP 401 counted as credential, so the other three classes fed the streak and its card told the user the digest was failing "for a non-credential reason" when it was entirely a credential problem — and transient blips, counted in a streak that only clears on a success, accumulated into the same false alarm. Invariants I7–I9 in daily-digest-credential-contract.md. The follow-up Q&A path uses the same routing without responseFormat (returns prose). On first run it loops from today - 7 forward so you get a backfill. Results land in SQLite via queries-daily-digest.ts with unread / archived / snoozed state and a per-digest conversation thread. The IPC surface lives in daily-digest-handlers.ts (list / get / mark-read / archive / snooze / follow-up / generate / unread-count / toggle-alert-dismissal / dismiss-all-alerts / list-suppressions / remove-suppression); the push events DAILY_DIGEST_GENERATED, DAILY_DIGEST_UNREAD_CHANGED, and DAILY_DIGEST_UPDATED keep the sidebar count and the open viewer live. Because that GENERATED push is fire-and-forget (the push bus has no replay), the briefing — and the weekly summary — also reconcile on return: a window focus / document-visible event refetches both, so a digest generated while Omniscio sat backgrounded overnight self-heals into the inbox the moment you come back to the window, with no manual refresh/reload/restart needed (useBriefingReconcileOnReturn, invariant briefing-reconcile-contract.md). There is no delete channel — archive flips the is_archived flag on the digest row, so digests are never hard-deleted from the database.
A day that cannot be built stops paying for itself (2026-10-16). Two defects were billing a customer's own API key all day. First, the parser recovered a JSON object only when the reply's code fence sat at the very first character — a model that answered with a fence preceded by "here is your digest" prose (or a fenced block with trailing prose) failed JSON.parse, so the day kept no digest at all. The parse now runs through the shared parseJsonFromLlmText (full-text parse, then a fenced block, then the first-{-to-last-} slice), so a fenced or prose-wrapped object is recovered and stored. Second, and worse: because generation is idempotent only on success, a day that produced nothing was re-generated on every 15-minute tick, forever, and the only brake was the F026 $1/day cost cap — which at ~16.9k tokens a call lands at roughly 45–50 calls, exactly the "45, 47 and 13 paid calls per day" a customer reported. A failing date now spends at most DIGEST_MAX_ATTEMPTS_PER_DAY (6) paid calls per calendar day, counted in the durable daily_digest_generation_attempts table so an app restart cannot reset it and windowed to the calendar day so a capped date never starves the days after it; the day then goes quiet at a verbose log instead of a warning. The count is cleared when a digest is produced. The cap gates the automatic tick and the boot catch-up only — Generate now is never blocked. A reply that carries no JSON object at all (prose, a fenced markdown block, an empty answer) is now treated as a FAILED attempt rather than stored as the day's digest: a stored non-digest would render forever with no regeneration, and it carries none of the alerts or pinned figures. That also means an empty reply finally feeds the "your digest keeps failing" streak, which it never did. Underneath all of it, the Anthropic backstop — the leg the digest lands on when the cheap provider is unavailable — now honours responseFormat: 'json_object': the Messages API has no response_format, so the reply is normalised instead (a fenced or prose-wrapped body comes back as the bare object, an already-bare object byte-identical, and a reply with no object untouched so a caller can still refuse it). One seam, so every JSON-mode caller inherits it. Invariant: daily-digest-generation-attempts-contract.md. Coverage: daily-digest-attempt-cap.test.ts · daily-digest-doGenerate-parse-fallback.test.ts · llm-provider-anthropic-json-mode.test.ts.
A retry loop stops paying, and a stranded day heals itself (2026-10-19). Three follow-ups after a customer reported a briefing retrying every 15 minutes for four days on their own API key. First, the JSON-only generation call now passes thinking: 'off': Haiku 5.5 thinks by default, and on the Anthropic leg those hidden tokens were billed and counted against the output budget — observed 4.8k–10.8k output tokens per briefing instead of ~2k–4k. The option is Anthropic-only at the provider seam, so the OpenRouter/Luna request is unchanged. Second, periodicCheck now also retries ONE stranded older date per hour (2–7 days back, newest first) — a date with no briefing at all, or only a basic ai-failed one — and it runs after yesterday's branch rather than behind its early returns, so it works on a day where yesterday was already built or had used its cap. It is bounded by the same durable 6/day cap plus two in-process bounds (a settled set for created/already-exists/no-activity, and a 6-per-day sweep count) for outcomes that never reach a paid call. Third, the 4th paid attempt for a date raises one deduped inbox card naming the date and the count, clearing itself when the briefing is saved; a timeout/abort/429/5xx stays out of the F113 give-up streak by design, so this is the only notice that loop produces. Underneath, a defect found while fixing it is closed too: the basic-briefing builder called .replace on an absent optional field, so a failed AI run often saved no fallback briefing at all — the same "never saved" symptom — and now always does. Invariants: daily-digest-generation-attempts-contract.md DA-I3b/DA-I3c/DA-I5. Release notes now name a background-helper model switch deterministically, including alias repoints (release-notes-curation-contract.md background-model-switches-are-named).
What the Alerts box contains (and what it deliberately excludes). The digest's Alerts section is reserved for non-obvious, cross-cutting problems you can't already see elsewhere — a broken build, a failing or wedged CI/cloud pipeline, an app crash (Sentry), or a systemic issue spanning many sessions. It does not re-list individual sessions that are errored / stalled / waiting / in recovery — those already appear in your inbox and the sidebar, so repeating them is noise (and a single session's transient status is usually stale by the time the morning digest is read). This is enforced two ways: the generation prompt tells the model to keep per-session attention in the action items section, not alerts; and a deterministic shared filter (filterInboxSessionAlerts in digest-data.ts) drops any alert whose text echoes an internal session marker (needs_you, wait_timeout, response_aborted, recovery_failed). The filter runs at generation, at render, AND in the share export, so even digests generated before this rule clean up the moment you open them — no regeneration needed. Genuinely useful infra/crash alerts carry none of those tokens and are untouched. Invariant: digest-alert-inbox-dedup-contract.md.
Sidebar bucketing (Active / Snoozed / Archived) — The active/snoozed/archived split is computed by a single pure helper, bucketDigests in digest-bucket.ts, called from DigestSidebarContent.tsx on every render with (digests, Date.now()). Active is anything with isArchived === false, snoozedUntil === null (or in the past), and digestDate within the last 7 days; the same 7-day cutoff folds older un-archived rows into Archived. Snooze re-bucketing is therefore passive — the row moves to Active on the first render after Date.now() > snoozedUntil, no timer or push event needed. The two flags live in daily_digests.is_archived INTEGER NOT NULL DEFAULT 0 and daily_digests.snoozed_until TEXT NULL (added in migration v107 — see incremental-migrations.ts and digest-sidebar-archived-migration.test.ts). Mutators are setDigestArchived(id, archived) and setDigestSnoozedUntil(id, iso | null) in queries-daily-digest.ts; the five archive trigger paths (middle-click, X button, right-click, E / Ctrl+W, and the open briefing's header button) all funnel into useDailyDigestStore.archiveDigest. The right-click context menu lives in DigestContextMenu.tsx; the row component (with its mobile-only X button + mousedown-based middle-click handler — auxclick doesn't fire reliably inside scrollable ancestors) is DigestSidebarRow.tsx. Snoozing reuses the generalized SnoozePalette.tsx modal (also used for sessions and emails) via a kind: 'digest' | 'session' | 'email' discriminator. The same setDigestSnoozedUntil mutator also drives the unified Inbox row — selectUnreadCount and digestInboxSelectItems in daily-digest-store.ts both filter out active snoozes so the inbox rows and the amber count badge mirror the sidebar's Snoozed bucketing without a separate code path. Individual briefing rows (2026-07-17): the inbox surfaces ONE row per unread briefing — each titled with its own date (formatDigestDate) and dismissed on its own — instead of a single collapsed "N new briefings" line; digestInboxSelectItems emits those per-briefing rows newest-first (plus a neutral-gray row for the briefing you're actively reading), and the backend digestToInboxItems projection mirrors it for the CLI /inbox view. The active-row highlight tracks the specific briefing via the digest store's selectedDigestId (threaded through the render surfaces as activeInboxDigestId) so only the briefing you're on highlights, never all of them. Right-clicking a digest in the Inbox opens the same DigestContextMenu; pressing S while the digest is the active inbox item dispatches the shared snooze-entity CustomEvent that the global SnoozePalette listener picks up.
Customization splits cleanly across the gather phase (which runs once per day) and the render phase (which runs every time you open a digest). Project exclusions are gather-phase: the service consults a getEffectiveProjectIds helper that subtracts the user's excluded-project list from the current real-project set before any session-summary work begins, so excluded projects never enter the prompt and never accrue Haiku tokens. Section visibility toggles are render-phase: the digest is generated with all sections present, and DigestViewer.tsx consults the section-visibility settings at display time to filter out hidden sections. This keeps "toggle on" instant with no regeneration needed, and a previously-generated digest immediately reflects the new visibility settings. Alert dismissals are persisted per-digest in SQLite (a JSON dismissed-IDs array on the digest row) and applied at render time by AlertsSection.tsx, which also exposes the per-row × button and the section-header "Dismiss all" chip. The customization UI lives in DailyDigestCustomizationCard.tsx, embedded in the existing settings page at DailyDigestSettings.tsx. The sidebar entry is DigestSidebarContent.tsx. Follow-up Q&A also uses Haiku so costs stay low even when you ask ten questions about the same digest.
Permanent alert suppressions (Phase 2) are a third customization track that hybridizes the gather and render phases. Each suppression is a row in digest_alert_suppressions (created by the toggle-alert-dismissal IPC when called with suppress: true) keyed by the {alertText, alertType} of the alert that triggered it. The generation prompt builder injects a SUPPRESSED ALERT TOPICS block listing every active suppression so Haiku is asked, in-prompt, not to re-surface those topics. Belt-and-braces: a post-generation safety filter (isAlertSuppressed + extractKeyTokens exported from daily-digest-service.ts) walks every alert in the model's response and drops any whose text shares ≥ 1 distinctive token with an active suppression — that catches the case where Haiku rewords the same topic. The suppressed-alert list and per-row × / overflow menu / hotkey wiring all share one identity check ({type, text}), exported as isAlertDismissed from AlertsSection.tsx so the dispatcher and the renderer always agree on what "the topmost undismissed alert" means.
Keyboard hotkeys (Phase 2) are wired through the standard keybindings registry — three actions in keybindings.ts (digestDismissAlert, digestSuppressAlert, digestDismissAllAlerts) all in the daily-digest category. The dispatcher in useKeyboardShortcuts.ts gates them: each keystroke is matched against a category-scoped lookup that only fires when activeProjectId === DAILY_DIGEST_PROJECT_ID, so X / Shift+X never collide with the same letters used in Gmail or diff-review. Handlers are registered from DigestViewer.tsx via the useShortcutAction hook (one call per action ID); each handler computes the topmost undismissed alert via the shared identity check and calls the corresponding store action (dismissAlert / dismissAlert({suppress: true}) / dismissAllAlerts). The Alerts section's hint chip uses the reactive useShortcutHint() hook (subscribed to the keybindings store) to render the live binding labels, so changing a binding in settings updates the chip text without a remount or page refresh.
Deterministic commit tally — DigestStats carries two fields, commitsTotal and commitProjectCount, that are injected into the parsed stats object after the LLM JSON is parsed — the model never authors these numbers, so they are deterministic and never hallucinated. gatherContext calls gatherDayCommits, which uses the same gatherCommitActivity() helper in commit-activity.ts over every project with ≥1 session on the local calendar day (date(started_at, 'localtime')). It reads all local branches rather than just HEAD, because Omniscio's worktree-per-session model lands commits on per-session branches the main checkout can't reach. Since 2026-10-01 it counts only authored commits — git rev-list --branches --since --until --min-parents=1 --max-parents=1 --format=…, then keeps a commit only when a person wrote it, it has exactly one parent (no merge, no parentless root), and it changed something (emptiness by tree oid, not --numstat, which measured ~297 s for a week against ~0.9 s for the tree compare). The parent filter and the machine-author test are the shared declarations in git-authored-history.mjs — the same ones behind the Weekly/Daily Contributor Report, so the two cannot disagree about those two questions. They are not the same number: the report additionally keeps only commits that touch code or tests, collapses rebased copies, and resolves its roster before the machine filter. Virtual-project sentinels are excluded via NOT GLOB '__*'. DigestViewer's StatsBar renders "N commits across M projects today" (hidden when zero).
Deterministic activity tally (2026-08-02) — alongside the commit tally, DigestStats carries three day-scoped activity counts — inboxArchived, userAnswers, and automatedActions — injected into the parsed stats object after the LLM JSON is parsed, so (like the commit tally) the model never authors them and they can't be hallucinated, and they add no prompt cost. gatherDayActivity in digest-context-gatherer.ts counts them app-wide over the digest's local day through the same off-thread reader: answered = your genuine replies to agents — an operator row that is not an auto-response, and is not agent-to-agent (metadata.peerMessage), not injected over the CLI server by another session or a cron (metadata.cliSourceSessionId / cliSourceCronJobId), not one of the app's own nudges (a crew-board notice metadata.boardNotice, the friction fixer's closing obligation metadata.frictionClosingObligation, or the phase-change to-do backstop metadata.phaseBackstop — subtracted 2026-10-09 after the 2026-10-08 briefing reported 5,360 "answered" of which 4,886 were these nudges, 474 real; a row queued over the CLI, metadata.cliOrigin, is subtracted too — an agent's whoever holds the key, T-194), and not the session's first operator row (an opening prompt is not a reply, whoever typed it). The "written by you" half is the SHARED predicate in human-operator-turn-sql.ts — the same one behind the Statistics panel's Your Messages card, so the two surfaces can never disagree about who "you" is; the day-windowed query that wraps it lives in digest-activity-sql.ts, which the SQL test imports rather than hand-copying. "Earlier operator row" is ordered by (timestamp, rowid), not timestamp alone — on a bare <, two turns written in the same millisecond would each find no predecessor and BOTH be discarded as openers. source='operator' alone is the channel, not the author — peer turns and CLI injections are written there with is_auto_response=0 deliberately, so the receiving agent treats them as a real new turn to answer; before 2026-09-09 the bare channel test made this chip report machinery as the user (measured on a real install: 2,426 "replies by you" for 2026-09-08, of which 1,443 were peer chatter and 252 machine-spawned openers — 73% machinery, corrected to 636). The Statistics panel's human bucket carried the same flaw and was fixed in the same change — its per-session denormalized message_count_* columns are re-derived by migration 20260909100102 (and again by 20261009050702 for the nudge exclusion), which clears every message_counts_synced_at stamp so the read falls back to a live scan while the background sweeper refills them; archived = inbox items you dismissed (inbox_alert_items.archived_by='user' — a new nullable actor column that tells a user dismissal apart from an automatic self-clear, since both stamp the same archived_at; every archive door now STATES its actor rather than relying on a default — an InboxArchiveActor (owner / agent / system; quiet is settable on the agent case alone) that resolveArchiveActor maps to both columns in one place. A person's own click records 'user' with no session; an agent or a route records a session (on a put-away made on the person's behalf it records 'user' with that session; a session retracting its own card records 'auto' with it, so it buys no quiet but is still attributable); only the machinery acting alone records a bare 'auto'. Locked by inbox-alert-contract.md invariant I28 and the repo-wide guard tests/unit/lint/21/inbox-archive-actor-is-stated.test.ts, which fails the build on any archive call under src/main that omits its actor); automated = automation-rule runs that fired, succeeded, and did something (condition_matched=1 AND action_result='success' AND action_type != 'pass' — the same "effective automation" filter as readChannelActions in summary.ts) plus away-mode auto-replies to an agent (is_auto_response=1 AND metadata.autoRule). StatsBar renders the three as an "activity" chip row (answered / archived / automated), each hidden when absent or zero so old digests and quiet days stay clean. Directly beneath that chip row, a human-vs-automated summary line sums the human side (answered + archived) against automation and states the % automated (e.g. "Human vs. automated: 1,093 by you · 1,931 by automation (64% automated)") — shown only when BOTH the human side and automation had activity that day, so it always frames a real comparison; a one-sided day is left to the chips alone. SQL coverage: daily-digest-sql.test.ts; actor-tag coverage: inbox-archive.test.ts.
Day-scoped usage ledger (2026-10-03) — the token and cost figures above are now read from a per-day ledger, session_daily_usage, instead of from whole session rows. sessions.input_tokens / output_tokens are cumulative counters with no day baseline, so the old read — sum each session over a started_at window — answered the day's question wrongly in both directions at once: it dropped every session that started earlier (which on a real day is the handful of long-running agents doing most of the work) and credited every session that did start in the window its entire lifetime. Measured on a real install for 2026-10-02, the briefing published 87.5 M tokens for a day whose transcripts carry 3.83 B, and $45.02 for a day that cost about $197. The ledger row is written on every turn, in the same statement as the session's cumulative counter, so the day's delta and the lifetime total can never disagree; it is keyed on the local day the work happened, so a session counts on each day it worked however long ago it started. It carries no history before it shipped — the per-day split was never recorded, so a backfill would be a guess — and an earlier day reads zero rather than a reconstructed lifetime total. Invariant: daily-digest-figures-contract.md a-day-figure-is-the-days-own-work. Coverage: session-daily-usage-ledger.test.ts.
The basic briefing and the exact numbers
- Every failure after the activity check saves a basic briefing.
doGenerateruns the AI path inside one guard; a failed outcome (no credential, cost cap, empty or unusable reply, provider error, truncation) saves a basic briefing from the gathered context, then reports the original outcome or error unchanged. Builder:src/main/services/daily-digest/digest-basic-briefing.ts. - A basic briefing is written only when the date has no row yet, so a no-key or cost-cap day that the 15-minute tick keeps re-checking stays as free and silent as before. It never resets the paid-attempt cap, the give-up streak or the gather cache.
- It is marked
basic: { reason }in the stored JSON (no-credential·cost-cap·ai-failed). The generator and the 15-minute tick treat a basic row as retryable; startup catch-up treats any row as present, so an older day is never re-billed. A successful AI run upserts the same row in place and also pushesDAILY_DIGEST_UPDATED, because the store ignores aDAILY_DIGEST_GENERATEDfor an id it already holds and an open viewer would keep painting the basic version. A basic briefing is never emailed (its message snippets are raw, not AI-filtered); the AI version is, once. Calendar times go throughformatCalendarEventTime, so an all-day event reads "All day" instead of a time-zone-shifted clock time. - Message text is scrubbed field by field (
scrubSensitiveData) and trimmed to a short snippet before it enters a basic briefing, because the stored summary is later re-sent to the follow-up model, emailed and shareable. - An email row in a basic briefing carries the sender and subject only. Its
summaryis empty: without the AI there is no summary to show, and falling back to the raw body put up to 160 characters of mail into the stored digest — and from there into every Overseer heartbeat brief (2026-11-04, T-1856). - One privacy seam scrubs the gathered context. The digest sends the assembled context to the model host both as the generation prompt and as the stored
contextJsonthe follow-up path re-sends, sogatherDigestContextWithReaderscrubs every string at its return:scrubObjectmasks emails, phones, SSNs and IPs by value pattern and contact fields by key name (names, subjects and channel names stay readable), thenmaskSecretShapedTokensDeepmasks unlabelled password-shaped tokens. The basic briefing is built from the same scrubbed context. - A password-shaped token is masked even with no label or vendor prefix.
maskSecretShapedTokens(secret-detection.ts) splits text on whitespace and quotes, strips edge punctuation, and replaces a core of 12+ printable-ASCII characters containing a lowercase letter, an uppercase letter and a digit with[SECRET_REDACTED]— unless it is a URL (://), a path (/…,C:\…,C:/…) or word-shaped (a run of 4+ lowercase letters with a character-class switch rate under 0.45). Measured on 20,000 random samples it catches about 90% of random 16-character alphanumeric tokens and about 98% of ones with symbols; words, markdown, identifiers, SHAs, UUIDs, model ids, timestamps, money and non-ASCII text pass untouched. A digit-less or word-based password is the accepted gap. - Rows stored before the fix are masked where they are re-read. The follow-up path masks the stored
contextJsonandsummaryMdit re-sends, and the Overseer fleet context masks the digest and weekly text before its character cap, so a capped token never leaves a readable head. Stored rows themselves, and the public share of an old row, are not rewritten. - The numbers are pinned, never trusted from the model.
pinDeterministicFieldswrites the headline stats (sessions, projects, total and out-of-pocket cost, the token split, commits, the activity tally) and the per-project session counts and cost from the gathered context, for the AI and the basic path alike. The prompt shows only a bounded slice of sessions, so a model-written count would never add up to the headline. A model reply that dropsstatsgets an empty object first, so the pins always land.tests/unit/lint/1f/digest-model-schema-numbers-are-pinned.test.tsreads the service for these assignments. - The cleared-for-you list is pinned into the briefing too and is no longer sent to the model, so it can no longer turn into one long paragraph inside a project card.
Related
- Briefings (part 1) — what a briefing is and how you read one.
- weekly-summary.md — the Monday-morning version of the same recap.
- gmail-integration.md — one of the most useful sources the digest draws on.
Last verified 2026-10-19